#but the automation is not the problem here it's that our livelihoods depend on things not being automated

[sillimancer]

I can't find it and don't remember who said it but I saw something where op was comparing liberals' reaction to hearing AI the same way conservatives react to hearing pronouns and I'm gonna be thinking about that all night now

[not-terezi-pyrope]

Hi! I read a lot of your posts on AI and I’m just curious about your take on the impact of AI on people and culture? I don’t use twitter so I don’t know what is being argued with AI, but my personal qualm is that it could obscure human experience. I’m not really concerned about the ownership of publicly available media, but I am worried about how AI can be used on a large scale by corporations to control human narratives. We already live in a world controlled by media monopolies, almost everything is owned by Disney, and their stuff is absolutely formulaic enough to be written by AI. While humans are borrowing just as much cultural information, one is still an original thought about how art should be created and presented. A corporation’s use of the technology will prime whatever is released for pure profit. It doesn’t need to be thoughtful at all. Again, this is already happening, but AI is just another way to streamline it. That’s part of why the writers strikes are happening right now. That kind of automation technology can shoehorn us into the necessary ends that corporations find most profitable. After reading your posts I realized that I was clumping in the technology with the ill intent of corporations. I wasn’t really considering all of its applications in its current form and I was kind of focusing on fear. You’ve changed my mind about some things. But human learning is dictated by modeling. We are sponges constantly absorbing the media shown to us. What if what’s available to us is always something easily automated by a company with its own interest? Avoiding that media completely would just result in personal isolation. I’d love to hear your thoughts on this as someone who understands the technology!

I think the key point here is one you've already made; corporations are already churning out mass media pulp optimized for profit, and have been doing so for years. At present, that pulp is probably more profitable when produced by humans, at least in terms of the overall story design of our media, so incorporating AI probably won't change much about its quality except for noticeable shortcuts that are deemed acceptable. But this is already true in digital productions. Corporations will probably still try to compete with a baseline of quality to make big budget releases acceptable to markets. You can't cut all the corners with cheap tech, which is why we are not seeing those shitty knock-off animated movies in cinemas. Things need to be a least competent even if they are schlock.

In the long term, if AI really is able to produce entire movies or similar that are comparable to human-made mass blockbusters, then the implication of that kind of depends on who is in control of the AI. It might lower the barrier to creating good looking movies, it might just let Disney make more animated princess flicks faster. I don't think either of those outcomes is essentially bad. There's good and bad ways it could manifest.

Specific failings of the technology could result in things like subtle bias, or feedback syndrome where the AI has little new content to work on as all new artworks are themselves AI generated. These are both technology issues that can be addressed, though, and are far enough off at the moment that it's hard to speculate about any real harms.

Honestly, the biggest negative impact currently threatened by AI is the obvious one that the writers are striking over; capitalism isn't set up to provide for human livelihoods under increased automation. The only solutions to this problem are strict regulation on how corporations can employ automation, or, perhaps more simply, watering down or abolishing capitalism. UBI is most realistic avenue for this at present, I think, even if it's not totally ideal (I'd like those corporations to not exist in a form that creates a profit motive, thanks).

Cultural concerns aren't nearly as scary a prospect on the horizon for me. Yeah, AI can be used to produce cheap schlock, but humans can already do that, can even automate it with free software somewhat with no machine learning required. As long as humans are ultimately deciding how and when the cheap schlock is made, as they are now, I doubt much will change about the essential character of popular art. This is all just my personal opinion, though. It's hard to predict exactly how things will shake out.

Thanks for asking!

[un-enfant-immature]

Tractable is applying AI to accident and disaster appraisal

“Happy to spend 10 minutes on our vision and the journey we’re on, but then, really, 15 minutes on what we’ve got today, what it is we’ve achieved, what it is our AI does,” says Tractable co-founder and CEO Alexandre Dalyac when I video called him a couple of weeks ago. “You can probably speed up all of that,” I quip back.

The resulting conversation, lasting well over an hour, spanned all of the above and more, including what is required to build a successful AI business and why he and his team think they can help prevent another “AI winter.”

Founded in 2014 by Dalyac, Adrien Cohen and Razvan Ranca after going through company builder Entrepreneur First, London-based Tractable is applying artificial intelligence to accident and disaster recovery. Specifically, through the use of deep learning to automate visual damage appraisal, and therefore help speed up insurance payouts and access to other types of financial aid.

Our AI has already been trained on tens of millions of these cases, so that’s a perfect case of us already having distilled thousands of people’s work experience Alexandre Dalyac

Dalyac launches into what is clearly a well-rehearsed and evidently polished pitch. “We are on a journey to help the world recover faster from accidents and disasters. Our belief is that when accidents and disasters hit, the response could be 10 times faster thanks to AI. So what we mean there is, everything from road accidents, burst piping to large-scale floods and hurricane. Whenever any of these things happen, things get damaged.”

Those things, he says, broadly break down into cars, homes and crops, roughly equating to $1 trillion in damage each year. But, perhaps more importantly, livelihoods get impacted.

“If a car gets damaged, mobility is reduced. If a home gets damaged, shelter is reduced. And if crops get damaged, food is reduced. Across all of those accidents and disasters, we’re talking hundreds of millions of lives affected.”

It is here where a little lateral (and non-artificial) thinking is required. Accident and disaster recovery starts with visual damage appraisal: look at the damage, say how much it’s going to cost, unlock the funds and rebuild. The problem (and Tractable’s opportunity) is that having an appraiser look at a car, house or field can take days to weeks depending on availability — and therefore so can accessing funds to start rebuilding — whereas the claim is that computer vision and AI technology can potentially do the same job in minutes.

“When you assess, that is basically a very powerful but very narrow visual task, which is, look at the damage, how much is it gonna cost? Today, as you can imagine, these kind of assessments are manual. And they take days to weeks. And so you instantly know that with AI that can be 10 times faster,” says Dalyac.

“In some sense this is a perfect class of AI tasks, because it’s very heavy on image classification. And image classification is a task where AI can surpass human performance as of this decade. If you have instant appraisal, that means faster recovery. Hence the mission.”

Dalyac says that part of Tractable’s secret sauce is in the many millions of proprietary labels the company has produced. This has been aided by its patented “interactive machine learning technology,” which allows it to label images faster and cheaper than typical labeling services.

The team’s focus to date has been to train its AI to understand car damage, technology it has already deployed in six countries, seeing the startup work primarily with insurers.

Related to this I’m shown a simple demo of Tractable’s car damage appraisal tool. Dalyac opens a folder of car images on his laptop and uploads them to the software. Within seconds, the AI has seemingly identified the different parts of the car and determined which parts can be repaired and which parts need to be entirely written-off and therefore replaced fully. Each has an AI-generated estimated cost.

It all happens within a matter of minutes, although I have no way of knowing how difficult the pre-determined and fully controlled task is. It’s also unclear how an AI can possibly do the full job of a human assessor based on a limited set of 2D images alone, and without the ability to peek under the hood or undertake further investigations.

“We’re trying to figure out how much damage there is to a vehicle based on photos,” explains Dalyac. “There’s some really tough correlations to pick out, which are: based on the photos of the outside, what’s the internal damage? When you’re a human you are going to have seen and torn down maybe about a thousand to two thousand cars in your whole life of 20 or 30 years of doing that. Our AI has already been trained on tens of millions of these cases, so that’s a perfect case of us already having distilled thousands of people’s work experience. That allows us to get hold of some very challenging correlations that humans just can’t do.”

You need to find real-world use cases that will make a difference, where you can surpass human performance Alexandre Dalyac

With that said, he does concede that a photo doesn’t always contain all of the necessary information, and that it might only have a certain level of accuracy. “You might need to then get a tear-down of the car and get photos of the internal damage. You might even want to get some data from the dashboard. And you can think that as cars get more sensors… the appraisal will be not just visual but also based on IoT data. But that doesn’t detract from the fact that we are convinced that it will be AI that will be doing this entirely.”

What is abundantly clear is Dalyac’s commitment to developing AI technology with real-world use that is commercially viable. If that doesn’t happen, he believes it won’t just be Tractable that will suffer, but the continued belief and investment in AI as a whole. Here, of course, he’s talking about the prospect of another so-called “AI winter,” citing a recent Crunchbase report that says funding for artificial intelligence companies in the U.S. has levelled off and even started to decline at seed stage.

“If you’re trying to make the $15 billion that has been invested into AI not fuck up and lead to something successful that will prevent an AI winter that will lead to continuous improvement, you need a really good return on that asset class. And for that you need those businesses to be successful.

“To make an AI company successful, really successful — not just an acqui-hire, not just an IP exit but a real commercial success that’s going to prevent an AI winter — you need to find real-world use cases that will make a difference, where you can surpass human performance, where you can change the way things work,” he says.

The reference to acqui-hire or IP exit takes on more meaning when you consider that Tractable was in the same cohort at Entrepreneur First as Magic Pony Technology, the AI startup acquired by Twitter for up to $150 million for its image enhancing technology. And most recently, the team behind Bloomsbury AI, another EF company, was acqui-hired by Facebook for $20-30 million.

To ensure that Tractable can continue its mission of applying AI to accident and disaster recovery — and presumably not sell too early — the startup has closed $20 million in Series B investment in a round led by U.S. venture capital firm Insight Venture Partners. Existing investors, including Ignition Partners, Zetta Venture Partners, Acequia Capital and Plug and Play Ventures, also participated. The new capital is to be spent on accelerating growth, expanding its research and development and entering new markets.

(The Series B also included an additional $5 million in secondary funding, seeing some investors at least partially exit. I understand Tractable’s founders sold a relatively small number of shares as they were permitted to take money off the table. Dalyac declined to comment.)

As we wrap up our call, I note that all of Tractable’s main investors, not including EF, are from the U.S. — something Dalyac says was a deliberate decision after he discovered the gulf between European and U.S. valuations.

“That’s a shame, isn’t it?” I say with my European tech ecosystem hat on.

“It isn’t; it’s enormous exports for the U.K.,” says the Tractable CEO who is French-born but raised in the U.K. “We have, as of today, the vast majority of our headcount in London. The entire product team is in London. The entire R&D team is in London. But most of the revenue comes from the United States. We are making AI an export industry of the U.K.”

[darkmoondancer]

A promise & dA/griffia drama

The promise & an explanation

So, you might have no idea what this post is about, or have any context to understand it. If that’s the case, it’s okay. It’s not actually important then. This post is for the huge amount of my past, current, and future clients who have commissioned works from me for the Deviant Art, Art RolePlaying Game, known as Griffia. Structural changes to that game have left some of you being extorted by that game’s owner for more money. In that extortion some of the public relations spin by said owner has put some entirely false but very public blame on me. So firstly. If you have ever bought art from me for this game. Please know, that I deeply did and still do appreciate your business and regardless of any and all drama over there, you can always reach me here to support your purchase. You bought character art from me, if you need more art, I am committed to fulfilling it. I simply can’t do it for free though, but I’ll do my best to help you out. I’m sorry I can’t do more, but despite the “spin” that griffsnuff (owner of the game) has put on this debacle I am not at fault and simply can’t afford to indulge her. I am not abandoning you and please don’t take anything said about me in places I can no longer defend myself as truth. Please just come speak to me. Secondly. Below is the full text of my last letters to griffsnuff and the first in bad faith. It is both exculpatory and personal. It’s not exactly something I’m proud of, but I feel as both an act of self-care and of conscious, it is something I needed to put out in the most wide reaching manner I have. But, there is a character limit on Deviant Art, and I am not exactly a concise writer. So here it is. If you have no idea what any of this post is about, I’m sorry to hurl it upon you. Trust me, you’re better off not reading it. It’s mostly drama.

The Open Letter

I'm not going to beat around the bush about this, but you are very much cornering Qvi, as well as all of the rest of us GAs or ExGAs, and in addition to that you are dismissing Qvi's very valid concerns. (Replying to this comment chain: here)

You are functionally forcing us to re-upload art that we may already have hosted, (or given our permissions for your old MLs to host), or else hazard looking like we don't value our past customers if we don't. By limiting art to the original artist's gallery, you aren't providing any sort of option for people to get permission to rehost art, especially when it comes to reference sheets, batch uploads, deleted artwork, and variations or edited artwork! In this way, you are very much cornering us. You are strong arming us and using this as an excuse to rip off your most loyal players.

This could easily have been avoided if you had taken ANY sort of courtesy to contact us GAs/ExGAs before the VML went live, and make any effort to sort something out that worked for all parties involved.

Why were we not contacted in any form to try to find some sort of amiable solution? Full disclosure: I am pissed off about not being contacted. I am pissed about being fired in the GA purge when you told all of us that you would be contacting us(GAs) again in the new year. You didn't even have the courtesy to reach out individually and tell us 'thanks for your efforts but we aren't asking you back'. You chose to let us sit there with our hopes up for months on end in silence? That's a dick move. Now, when some of us have moved on, or are in the process of trying to, we are being contacted by past customers about a game we might not even play anymore (because it broke our trust and our hearts) all because you decided to put up new random rules on your website? A website that is actually displaying direct image links[1] instead of doing something sensible like coding in a pop-up permissions window that shifts the upload liability to the user instead of the site? You know? Like every modern website doing business inside the bounds of effective DMCA enforcement? Which, FYI, you clearly are not within. I'm not either, as a Canadian. Youtube and DeviantArt are, as corporations existing... INSIDE AMERICA! And don't mumble about WIPO, it's a freaking diplomatic tool, not an actively enforceable legal instrument! Neither is the DMCA, in practise. But that is way beyond the scope of this rant. Seeing as you'd need to have a basic understanding of the history surrounding DMCA and the ensuing voluntary policies put in place as a protective measure by the pre-tech bubble western tech hegemony against the insanity that is the actual base law. Which clearly evidenced by your own words and actions is leagues beyond you. That being said, I am not upset with my customers, or anyone who has contacted me in hopes of getting their art sorted out. I am all on-board to help them get things working, because I respect them and am grateful that they supported me in the past and bought my art. However, I am sick of the abuse that players, mods/exmods and GAs/ExGAs continue to endure in this community simply because you keep changing your product on a whim. Griff, these customers bought art that represents a character in your game from someone you granted license to do so. This art was expressly sold for use in your game, which includes being hosted on the game's ML. All art that GAs made for your game was sold to the customers, as a product that works within the World of Griffia. This was explicit by your own actions, words, declaration and by the license you implicitly extended to your agents. Your employees, the GAs! You know, the people you paid with licensing rights to do work for you? Licensing rights that had DIRECT monetary value! The people you also paid with access to monetary instruments known as proxy currencies, or by game speak, ITEMS! For fuck sake, you prattle on about DMCA while ignoring the actual laws you are beholden to? How in the hell have you not seen the rulings on digital goods that can be exchanged for real money?!? You are Norweigan! You are PART of the fucking European Economic Area. Those portions of EU laws explicitly apply to you! THEY ARE LAWS REGARDING ECONOMIC ACTIONS! How can you be so comfortable quoting laws with such blatant inaccuracy? And now, through all of this, you are telling your customers that it is not your problem? If they don't get to use the art they purchased with real money, in the context it was bought for, then they are clearly suffering a loss. Their character is being devalued by your actions unless they give YOU money? Their investment, as is the case with a great deal of players, is now worth less! They have no other option other than to suck it up? To deal with the n/a page or to pay a now select-few artists (your agents, your employees) real money to fix it? To give you or your agents money or suffer the consequences of a depreciated asset? And make no mistake, your "guest artists" are your agents. They are your employees. Just because you don't understand the most basic principles of money doesn't make it any less so. You are paying them in rights. Your most loyal fans (customers) now must shell out even more money to get a new rebase, or draw themselves new art that can't go in this official slot at all? The least you could do is allow people the option to make themselves a rebase on your official bases that can count as official art. 

On top of it all, you aren't actually providing credit to the original artist since you just have direct image links. Embedded images that only exist so long as dA continues to host that particular library file. Images that don't even link back to the artists original post, but steal it directly from the hosting infrastructure[1]!?! Hosting infrastructure that will clear eventually. Whether by an inactive user, scheduled maintenance, or malice, the links will die. Data costs real money. Somewhere in the world it is taking up warehouse space. Regardless of whether it's a digital set of bits on a storage drive. It costs actual money to host. Likely the real reason you don't want to host the images yourself? The DMCA explanation is gibberish, at best, as noted above. So no, dA isn't going to be hosting the current image files forever, and yes, you will inevitably have images go missing again, which means you are going to be right back at square one (missing images because they got deleted), which is what you've been trying to say you are attempting to avoid this whole time! Asking for a 'stable dA link' is literally nonsense. An entire point of having your own website for a game like this, is so that you are not dependent on dA hosting your stuff… and yet you are making this exact mistake again. You are making a mistake, while in the same breath, freaking out about that exact mistake? Are you insane? You are integrating automation and other features, but to have waited a whole year to see that this new VML doesn't even provide credit to the original artist (either by text, via link-through, or ANYTHING), and be greeted with arbitrary restrictions like no background, no pets, only individuals AND all has to be posted by only the original artist specifically only on dA galleries… AND to be followed up with dismissals that hosting this art is our own problem, when it's in fact all for YOUR game… and to support YOUR livelihood… I'm infuriated that I'm not seeing you taking any responsibility for this or respecting your customers for having put money into your game economy by buying a product that's value was solely granted by it's rights of use within your game! You keep breaking your promises, Griff, and there's only so much ground left to stand on.

I feel like you are not only being flippant towards us GAs/ExGAs, but trying to make this the GA's problem. Is this just so you can hide from your responsibility and milk even more money of your most loyal fans? Some of us aren't even a part of your team anymore and don't have any obligations to your game whatsoever. But you're trying to sell some bullshit to shift the blame regardless? You are breaking your promises to YOUR customers, who are trying to play in YOUR game! Before you go saying that I'm simply blaming all of this on you directly, Griffsnuff… I am. Griffsnuff, you are the owner of Griffia. I don't give a flying fuck what your godly levels of gibberish have to say otherwise. Especially since the words included in that gibberish changes with the seasons of your feelings. You ultimately carry all blame. Responsibility ultimately falls on… that's right… you! You are the owner. That is how reality works, and I agree it sucks, but that is the responsibility you take on when you decide to run a business. You can't keep telling everyone you are doing this all with the intent to have fun and make a game for you and your friends to play. No. If this was all for fun, you would not be selling characters… for large amounts of money… that explicitly say in them that they can be used in this game… and encouraging people to buy these characters to play in this game with you…. That is literally a business. (A proprietorship is still a type of business, so yes, even you as a singular artist are running a business when you sell anything!) So if you're upset that people are expecting to continue to use their characters within your world… who made up the rules that the characters could be sold into your world? You did. You allowed us GAs/ExGAs to freely make in any capacity, quantity and quality the species that you assigned us, as well as Rebases in literally any form/style/quality/etc we wanted. So long as they were on a transparent background, were solo, and at least vaguely followed (when you could be arsed to double check) the non-existent species information that you scribbled across three groups, countless stashes, and numerous google docs. Yes, eventually even you, started to realize that not giving us structure and real (consistent) written rules to follow was starting to backfire on you. It seems you even finally realized that you were giving away actual value. You were in fact using licensing as a proxy for cash. I think even you realized you were losing money out of your own pocket from your utter failures to comprehend your own decisions. But wait, this is the part where it's everyone else's fault right? Wrong. You gave out those license rights. You told people that they were buying a shiny special character design that could do things based on their status, subtype or species. Then you were surprised that people were upset that you made yourself out to be a liar when you changed those things to have less value? You allowed GAs to make as many of their designs as they wanted… and then act surprised when the market tanked because suddenly there was more supply than demand for any given thing? You create new and exciting things with nothing more than whimsy and then act defensively when people get upset with you when you randomly alter your half baked ideas?

You have created all of this for yourself. 

TL;DR: You have clearly, and repeatedly, over many years shown a complete inability to comprehend markets, or market forces. Regardless of creed, philosophy, or ideology the inscrutable fact is that Griffia is first and foremost a market. Griffia is not art. Griffia is not stories. Griffia is not a game. It is a marketplace in which people (yourself included) create stories, play a game, and trade art. Trade makes a market. Gravity makes apples fall. The sun makes life possible. Your job, your profession, and your livelihood is the production and management of this marketplace. The ramblings in which you lament otherwise are no defense from reality. You sold a marketplace, to sell your art. You hired artists to work for you and paid them in licensing rights. Your incompetence left buckets of money on the table, and when you finally noticed it, your reaction is to fuck over every one of your customers who didn't predict the future collapse of your idiocy? Fuck you, Griffsnuff. 

I'm done giving you my attention, my effort, my work, and all the space you have rented in my brain over the last 3 years. I would have been better off dedicating that time to literally anything else. I only have myself to blame for sticking around for so long and continuing to hope that maybe things would improve. 2 years ago, I told you all of this in less jargon and kinder words[2] because I wanted to help you grow. A year ago, I despaired and wrote you a second letter[3], urging you to reconsider your actions and warning you that you were leaping off a cliff-face. Today, 3rd time is the charm and I'm fucking done. This letter was for me, not for you. I'm so fucking done. This is where I draw my line. I hope for their sake, that others do so too- players, mods/exmods and GAs/ExGAs alike. They deserve to be treated with respect, not whatever this web of constant lies, broken promises, ass-kissing, randomly enforced unwritten rules, favouritism, and arbitrary yet crushing lore/anatomy changes is pretending to be. This is my third and final letter I will write to you Griff, as I fear it yet again will fall on deaf ears as my second one did. I sincerely hope you will read it, and take the time to thoroughly understand it. But I doubt it. Past clients, I'm still supporting you. I'm only done with Griff and Griffia, not with you. Check my blog here for the preamble to this letter where I address this.

PS: I would like you to answer 'one' question. If I choose to leave dA entirely to get away from this mess of a game once and for all, how do you propose to deal with the artwork I made for others? How will you honour that agreement? The agreement in which you licensed me to make art for your customers in exchange for services rendered. How will you cope with your resentment that you lost a bunch of money because you are shitty at your fucking job? The one where you run a marketplace for art governed by licenses you don't understand? It certainly wouldn't be fair to punish your customers by taking away their access to the art they bought from me. But, that seems to be your current choice? You have currently cornered everyone and provided no solution for those who are affected. If I were to simply go through and delete all my artwork or account to get a fresh start what would happen? If granting them permission to rehost my artwork isn't enough, and granting your past MLs permission to do so isn't enough, and granting your current VML permission to use any and all Griffian art I have made up until today isn't enough… then I'm at a loss. Is your completely mindless understanding of a set of laws that doesn't affect you, enforced by a policy guideline voluntarily adhered to, and failed to be managed by you due to sheer ignorance, going to stop you from doing what's right? So my 'one' question?

Are you really willing to be an extortionist? Because, despite your massive inability to comprehend reality, I suspect at the very least you can wrap your head around the facts. If you continue this tact; You are simply and plainly extorting your most loyal players while lying to their faces about your intentions and covering it all up with blatant defamation of your past artists. You are telling people that if they don't give YOU money, the things they own, will be worth less. By any and all moral definitions, that is extortion.

PPS: Now legally? Well holy shit that is WAY more scary seeing as how you are Norwegian and are not as protected by the disgusting monster that is crony capitalism. You NEED to see a Norwegian corporate lawyer regarding your legal requirements of disclosure. Because, according to your own nations Criminal Code (as translated to English) "Chapter 25. Extortion and Robbery", if you are altering the marketplace in an unlawful manner then you are using "unlawful conduct" as a tool to cause another party to suffer "loss". The unlawful conduct being your actions and the loss being the devaluation of artless characters in a market (in your words "a game") about character art. You could be facing your nation's minimum sentence of prison time for extortion. How strong are your consumer advocacy groups? How focused are your cops on white collar crime? How much do you trust your law enforcement to protect individuals over business interests? Morally speaking you are clearly extorting people, but you need a lawyer to give you advice on whether you are doing so criminally. You absolutely fucking need to go see a lawyer! In your own country!! To see if you are even allowed to abruptly change a market in this way. Your nation will have its own complex legislation prescribing your conduct as a market provider. It will likely fall under contract law. PPS:PS: Please keep in mind, when you do your due diligence (reading your own countries laws) before doing the smart thing and talking to a lawyer, that in legal writings, subsequent statements and sections are subservient to the prior. Seriously, go see a lawyer. You could be in actual real danger. The minimum sentence for extortion is jail time with the option of a fine in addition! (Not even touching the whole loot box/winning flower/treasure bean stuff that you are laughably under protected from if other EU web presences are to be used as an example.)

I'm not an expert, but I can read. You should too. Then pay a fucking expert.

[1] Inspect the source and URL, WIXMP is Wix Media Platform, the backbone of the DA hosting architecture. They switched to it while ramping over for the move to Eclipse. [2] The first letter, written in good faith.

[3] The second letter, still in good faith, but much more upset.

[wellplacedrocket]

Sidequest Log | Technology is (probably) not gonna replace your bartender, but it’s still pretty damn cool

When I’m at a bar having a beer, the tap handles tease me. They make them so ornate and visually interesting now, and each beer has its own brewery-designed handle. They are works of art unto themselves; it’s like a statue park, except each statue represents something interesting and delicious going into my mouth. I fantasize about being my own bartender, expertly and with flourish pouring out my own glorious nectar of the gods into a tilted pintglass, knowing just when to stop so it doesn’t foam over.

I didn’t know that anyone had ever packaged my longing into a business plan, but this is 2018, so I guess that should have been obvious. On a whim recently, I impulsively stopped in Tap Society on 46th St. and Grand Ave in Minneapolis. I expected a typical burger bar with ESPN on TVs, and instead I spent the next hour or so getting my head around an entirely new (to me) business model: a self-serve bar.

I stuttered out my stupefied captivation with this idea to General Manager Michael Nienaltowski. What wondrous Narnia have I walked into here? He replied “Yeah. So obviously as you said, the tap wall, the self-serve tap wall, is really the star of the show here.”

It was very kind him to coach me up, because even though Tap Society is super simple, I’m so used to the standard bar procedure that I felt like I was fumbling every step of my whole be-my-own-bartender fantasy.

“We were looking up there and when we first came in, I thought that was a bar,” said Josh Peterson of Minneapolis, indicating the long table next to the tap wall. “There was someone over there and we thought that's the bartender. Then I realized. So once we figured it out, it's kind of cool.”

“I like that you pour your own beer, I think it’s fun,” added Tina Dunlap, also of Minneapolis.

When you enter, you “check in” at a station near the door and they scan your debit/credit card, hand it back to you, and you get another card from the bar - your Tap Card. Then you walk up to the tap wall and grab a glass.

Each pair of taps has a small screen above it, and then the beers’ names, styles, ABV, and cost per pint are posted on larger TV screens a few feet above. You insert your Tap Card into a slot, which unlocks the tap, then stick your glass under the spout, and fulfill your self-bartending fantasies. You can see the price of your pour accrue on the small screen above your tap, like you’re at the gas pump. It tallies that charge to your Tap Card. When you’re ready to be done, you turn your Tap Card in to a human or drop it down a box. Your name has already been associated with that Tap Card, so they read it from there and know what to charge your debit/credit card. No cash used at all.

So let’s pros/cons this thing! First of all, you can pour as small a glass as you want, and you’re only charged by how much your pour. They had a Märzen that I wanted to try just a bit of, so I got a pour that was about $1.70 - a few mouthfuls. Anticipating the draw that sampling would be with their tap wall, they have plenty of small glasses available. You can easily make your own flights. “I've had many customers come up with smaller glasses and pour six or seven different beers and it costs less than 10 bucks depending on what you're drinking,” Nienaltowski said.

Secondly, there’s no one to tip. Food is ordered from kiosks, and you’re getting your drinks yourself. Floor staff consists only of a few customer service people to help coach customers up as needed or answer questions, and “food runners” who bring it out to you. I don’t want to get into virtue signalling about my tipping habits, but I’m aware that bars and restaurants have carved out embarrassing minimum wage exceptions and expect their staff to be almost entirely paid in tips. When pressed a little, Nienaltowski said they pay “competitively,” and I wasn’t sure what that meant. I tend to be very skeptical of business posture toward labor, no matter how nice the businessperson is. But later, I was assured by a kitchen guy out on break listening in that he’s pretty happy with his wages. Net result for me was my total tab was noticeably lower than it usually is, thanks to short pours and no tip.

Now I’ll strawman up some downsides. First and foremost, if you’re into cocktails, you’re SOL. There’s no one to make them for you, so they don’t even bother. I suppose they could premix some gin and tonic or something and put it on tap if they really want, but eh. Literally the only time I ever order cocktails at bars is if I’m there because it’s specifically a cocktail bar and I’m splurging on fancy. I will concede that lowball glasses full of lime wedges and rows of weird bottles with pourers in interesting arrangements behind the bar are more visually interesting than Tap Society’s tap wall, which is all digital and kinda antiseptic.

On a related note, if you go to bars specifically for bartenders, Tap Society again strikes out. Bartenders have a romanticized vocation in our culture, and I get it. I’ve had a great time chatting to bartenders, the job tends to attract those who are comfortable schmoozing and multitasking often demands people of quick wit and efficient focus. I’m aware they’re at their job and therefore a captive audience, so I never want to dominate a bartender’s attention. But a good bartender can cheer you up, get a laugh, blow your mind, and make your night.

I wouldn’t want all bars to ditch their bartenders, and they won’t. But I think there’s room in the economy for tap walls, and I was having too much fun playing with the taps to miss the person who’s usually pouring them for me. And if it’s really busy, it’s far more efficient to let each customer come and line up at the tap they want (if there even is a line) than to create the customer service choke point of 25 people mashed against the bar all politely jockeying for elbow space while 4 bartenders hustle with little sense of any queue order.

Bartenders reading might think this tap wall is the enemy, here to put them out of a job. When I texted my dad about what I’d discovered, his response was “No bartenders. More unemployment.” Fair point! Is Tap Society the vanguard of the next wave of the Tech Priests of Silicon Valley, here to sacrifice real human livelihoods at the alter of technological efficiency?

“I think instead of looking at we are saying we have less people, I think it's the way we utilize the staff and when we have them on,” says Nienaltowski. He figured when it’s slow, they probably have the same staff size as they would if it were a normal sitdown bar/restaurant, but fewer staff compared to the standard model during busy hours. “I don't think it's necessarily looking at it as we cut bodies for automation to save a dollar, it's just the way our model works. It doesn't require all those people. And the people that we have, we utilize their strengths in different ways.”

It was unfair of me to try to make him speak for the entire socioeconomic footprint of technological advancement. He’s defending the employer and focusing just on how Tap Society works. But my dad’s grumble still holds, obviously. Tap Society is a medium-sized establishment, and it just isn’t going to hire as many total workers as it would if it were a normal bar. They’re obviously not devoid of the human touch in there. It was slow, and I chatted with humans the whole time I wasn’t taking pictures. I don’t know what the solution is to automation’s effect on the economy. Some kind of national job retraining and relocation assistance program, I guess.

Here’s a definite downside: no happy hour! I feel they could easily program the tap wall to drop the prices during certain hours. Not sure why they don’t.

You also might thinking that it would be super easy for all the local drunks to post up and knock back Miller Lite after Miller Lite with no gatekeeper keeping them from getting shitty. Valid concern. The Tap Cards only allow so much pouring before the card locks and you have to check in with a human to unlock it again. I’m not sure if it keeps track of volume or total tab applied to the card or maybe even alcohol consumed, tracked by ABV. It’s not a big deal to get your card reset, BUT it would be a problem if you were trying to buy a round for your table. I was at about $16.00 when it locked up on me, and I was done anyway, but it would fill up the card quick if there were other people involved.

I don’t think places like Tap Society are ever going to replace all your favorite bars and bartenders. But I stopped myself when I was about to type “I don’t think it’s going to revolutionize the bar/restaurant industry.” I’m not so sure about that, when I think about it. The tech is really, really cool, my friends. And as someone for whom beer accounts for 99% of my bar selections, it’s a simple, fun, interactive, and overall less expensive way to deliver it to me, with plenty of room for a tantalizing selection. If you like going to bars and drinking beers a lot, Tap Society and places like it might be perfect for you. And if you’re worried it’s another digital interloper trying to supplant a sacred analog ritual, that’s probably not going to happen. Probably.

“I think you can have both,” Josh told me. “I don't think this is the way where the world is going to go, but having a couple of these around, I don't think would hurt anything. And it’s cool, it’s a concept.”

“The problem is, you just always need a human,” said Ryan Lagieski, also of Minneapolis. “We’re not to the point where everything can be automated."

[theconservativebrief]

Last year, in the New Yorker’s dive into whether the sharing economy would persist, writer Nathan Heller compared the gig economy to the “Conscious III” people from Charles A. Reich’s 1970 book The Greening of America who “simply do not imagine a career along the old vertical lines.” The gig economy aligns well with the desire for flexibility and pushback on traditional work hours.

Though the share of the population participating in it varies depending on how you define the term — from less than 5 percent for strictly online platforms like Wag and Airbnb to 40 percent for any type of freelance or contract work — the consensus is that those numbers will continue to grow. So how are all these gig workers faring financially?

A new JPMorgan study looked specifically at how the online platform segment of the gig economy has changed since 2013 by examining transactions between Chase bank accounts and 128 online gig platforms. The study found that those who are using apps and sites to lease their assets, whether an apartment or a car or a parking space, saw a 69 percent increase in income.

But those in the transportation sectors saw an increase in the number of platform users and a significant decrease in income. Monthly wages fell 53 percent from 2013 to 2017, and the average driver’s income decreased from $1,469 to $783 per month. In those years the number of employees on transportation service platforms also increased — now 2.5 percent of those working in the online platform economy are working in transportation services.

Why people turn to gig work has a lot to do with how old they are. According to a 2018 study, Gen X-ers work the most hours per week of any generation and usually have a high school education. They are also more likely to rely exclusively on gig work for income. Baby boomers usually have high degrees and are retired. They use on-demand service platforms to make money they may not be getting in retirement. Millennials use it as a side gig way to pay off debt and are often more highly educated as well. Both millennials and boomers say they enjoy the flexibility offered by gig economy jobs.

But the freedom so often marketed with gig economy jobs is not as liberal as it seems. Cases about whether drivers are considered full-time employees or contractors have plagued online platform companies across industries. This year, a Grubhub driver brought a case against the food delivery platform, saying it controlled a significant amount of his time the way a manager controls employee shifts, and therefore, he should be offered minimum wage, benefits, overtime pay, and reimbursement for expenses. The court disagreed and said Grubhub drivers are independent contractors, relieving Grubhub of providing any full-time employee benefits. Uber has also faced suits from drivers saying the algorithm Uber uses exerts control of a driver’s schedule, but no court has ruled for drivers to be treated as full-time employees.

So will the gig economy eventually overtake traditional employment? Does this decrease in wages and increase in participants foreshadow an even more precarious future for all gig economy workers? I asked four sharing-economy experts for their views on the current state of the gig economy and where it might be headed.

Paris Marx, author of Freedom From Jobs: How Automation Will Revolutionize the Future of Work

Gig economy jobs will not overtake full-time jobs, nor was it ever likely for that to happen. The growth in the number of people engaging in the gig economy, as reported by JPMorgan, is not a hopeful statistic, but an incredibly worrying one. It does not show the growth of a positive new employment sector, but illustrates the growing precarity of the American people.

But if things continue as they’ve been, with wages continuing to stagnate (or grow very little), health insurance getting more expensive, and the rich capturing the bulk of economic gains, then yes, the gig economy will keep growing because desperate people will seek out some way to earn extra income and the barrier to entry is often lower in the gig economy than in traditional employment.

However, if proper living wages are instituted, Medicare-for-all is adopted, and taxes are hiked on rich people and corporations to reinvest in infrastructure renewal and a Green New Deal — programs which could create millions of well-paid jobs — people would be less likely to turn to the gig economy and there would be greater pressure to increase the labor standards that those companies have to abide by.

Keith Ryu, founder of the online hiring platform Fountain

This entrepreneurial spirit is what is driving this new, flexible workforce, or new service economy. We also see the budding proliferation of the on-demand business model that is built on the very nature of people wanting immediate access to goods and services from the convenience of their smartphone.

The only gap in all this is the way these companies recruit and hire this dynamic and changing workforce. Companies have spent resources in optimizing services to their consumers but have not put this same investment in their hiring practices. For this new service economy, barriers need to be removed from the hiring process.

The biggest one is the sourcing and hiring process. The easier companies make it for candidates to apply, the faster they can meet their workforce demands. For the candidate, having this effortless access to immediate work makes the success of their livelihood possible.

The gig economy is expanding, companies are adapting to this growth, and the result will be that companies who make the hiring process easiest, will win over the most qualified workers. Companies, however, that continue to look for gig workers with outdated methods and tools will continuously lose the battle for candidates and become stagnant.

Ryan Nunn, policy director at the Hamilton Project, an economic policy initiative at the Brookings Institution

In principle, one real advantage of the gig economy is that workers can use it on a temporary basis when they face shortfalls in income from their primary employment, or when they experience unexpected expenditures. But on the other hand, we should be aware that people working primarily or exclusively in the gig economy generally lack labor protections (e.g., collective bargaining rights) and non-wage benefits (e.g., health insurance) that are more often associated with traditional employment.

The technologies and business practices that make the gig economy possible are still relatively new, and it seems likely that entrepreneurs will find more opportunities to expand the gig economy. I think it remains to be seen whether that expansion will result in any large shift in workers’ primary employment.

Diane Mulcahy, author of The Gig Economy

Low-skilled workers who are poorly paid and don’t have access to benefits is a problem of our economy and our society, and it’s not specific to the gig economy. That Uber driver is actually worse off in the traditional economy.

The problems of lower-skilled workers not being paid and not having access to benefits, that’s a systemic problem. I think if you’re looking at Uber driver or a lower-skilled worker, what are the current alternatives? In a traditional job economy, they could be working at a fast food restaurant or retail, and under those options their economic situation is similar. The difference is when they’re an Uber driver, they can control how much they work. They can control their life. In the traditional jobs economy, they are given shifts.

The gig economy is here to stay, and it’s only growing. People want to work differently. But as long as we have a labor market which penalizes you if you are not a full-time employee, I think it’s unlikely that the gig economy will be the predominant way of working.

Original Source -> The gig economy isn’t going anywhere. 4 experts explain why.

via The Conservative Brief

[thefeedpost]

Tractable is applying AI to accident and disaster appraisal

“Happy to spend 10 minutes on our vision and the journey we’re on, but then, really, 15 minutes on what we’ve got today, what it is we’ve achieved, what it is our AI does,” says Tractable co-founder and CEO Alexandre Dalyac when I video called him a couple of weeks ago. “You can probably speed up all of that,” I quip back.

The resulting conversation, lasting well over an hour, spanned all of the above and more, including what is required to build a successful AI business and why he and his team think they can help prevent another “AI winter.”

Founded in 2014 by Dalyac, Adrien Cohen and Razvan Ranca after going through company builder Entrepreneur First, London-based Tractable is applying artificial intelligence to accident and disaster recovery. Specifically, through the use of deep learning to automate visual damage appraisal, and therefore help speed up insurance payouts and access to other types of financial aid.

Our AI has already been trained on tens of millions of these cases, so that’s a perfect case of us already having distilled thousands of people’s work experience Alexandre Dalyac

Dalyac launches into what is clearly a well-rehearsed and evidently polished pitch. “We are on a journey to help the world recover faster from accidents and disasters. Our belief is that when accidents and disasters hit, the response could be 10 times faster thanks to AI. So what we mean there is, everything from road accidents, burst piping to large-scale floods and hurricane. Whenever any of these things happen, things get damaged.”

Those things, he says, broadly break down into cars, homes and crops, roughly equating to $1 trillion in damage each year. But, perhaps more importantly, livelihoods get impacted.

“If a car gets damaged, mobility is reduced. If a home gets damaged, shelter is reduced. And if crops get damaged, food is reduced. Across all of those accidents and disasters, we’re talking hundreds of millions of lives affected.”

It is here where a little lateral (and non-artificial) thinking is required. Accident and disaster recovery starts with visual damage appraisal: look at the damage, say how much it’s going to cost, unlock the funds and rebuild. The problem (and Tractable’s opportunity) is that having an appraiser look at a car, house or field can take days to weeks depending on availability — and therefore so can accessing funds to start rebuilding — whereas the claim is that computer vision and AI technology can potentially do the same job in minutes.

“When you assess, that is basically a very powerful but very narrow visual task, which is, look at the damage, how much is it gonna cost? Today, as you can imagine, these kind of assessments are manual. And they take days to weeks. And so you instantly know that with AI that can be 10 times faster,” says Dalyac.

“In some sense this is a perfect class of AI tasks, because it’s very heavy on image classification. And image classification is a task where AI can surpass human performance as of this decade. If you have instant appraisal, that means faster recovery. Hence the mission.”

Dalyac says that part of Tractable’s secret sauce is in the many millions of proprietary labels the company has produced. This has been aided by its patented “interactive machine learning technology,” which allows it to label images faster and cheaper than typical labeling services.

The team’s focus to date has been to train its AI to understand car damage, technology it has already deployed in six countries, seeing the startup work primarily with insurers.

Related to this I’m shown a simple demo of Tractable’s car damage appraisal tool. Dalyac opens a folder of car images on his laptop and uploads them to the software. Within seconds, the AI has seemingly identified the different parts of the car and determined which parts can be repaired and which parts need to be entirely written-off and therefore replaced fully. Each has an AI-generated estimated cost.

It all happens within a matter of minutes, although I have no way of knowing how difficult the pre-determined and fully controlled task is. It’s also unclear how an AI can possibly do the full job of a human assessor based on a limited set of 2D images alone, and without the ability to peek under the hood or undertake further investigations.

“We’re trying to figure out how much damage there is to a vehicle based on photos,” explains Dalyac. “There’s some really tough correlations to pick out, which are: based on the photos of the outside, what’s the internal damage? When you’re a human you are going to have seen and torn down maybe about a thousand to two thousand cars in your whole life of 20 or 30 years of doing that. Our AI has already been trained on tens of millions of these cases, so that’s a perfect case of us already having distilled thousands of people’s work experience. That allows us to get hold of some very challenging correlations that humans just can’t do.”

You need to find real-world use cases that will make a difference, where you can surpass human performance Alexandre Dalyac

With that said, he does concede that a photo doesn’t always contain all of the necessary information, and that it might only have a certain level of accuracy. “You might need to then get a tear-down of the car and get photos of the internal damage. You might even want to get some data from the dashboard. And you can think that as cars get more sensors… the appraisal will be not just visual but also based on IoT data. But that doesn’t detract from the fact that we are convinced that it will be AI that will be doing this entirely.”

What is abundantly clear is Dalyac’s commitment to developing AI technology with real-world use that is commercially viable. If that doesn’t happen, he believes it won’t just be Tractable that will suffer, but the continued belief and investment in AI as a whole. Here, of course, he’s talking about the prospect of another so-called “AI winter,” citing a recent Crunchbase report that says funding for artificial intelligence companies in the U.S. has levelled off and even started to decline at seed stage.

“If you’re trying to make the $15 billion that has been invested into AI not fuck up and lead to something successful that will prevent an AI winter that will lead to continuous improvement, you need a really good return on that asset class. And for that you need those businesses to be successful.

“To make an AI company successful, really successful — not just an acqui-hire, not just an IP exit but a real commercial success that’s going to prevent an AI winter — you need to find real-world use cases that will make a difference, where you can surpass human performance, where you can change the way things work,” he says.

The reference to acqui-hire or IP exit takes on more meaning when you consider that Tractable was in the same cohort at Entrepreneur First as Magic Pony Technology, the AI startup acquired by Twitter for up to $150 million for its image enhancing technology. And most recently, the team behind Bloomsbury AI, another EF company, was acqui-hired by Facebook for $20-30 million.

To ensure that Tractable can continue its mission of applying AI to accident and disaster recovery — and presumably not sell too early — the startup has closed $20 million in Series B investment in a round led by U.S. venture capital firm Insight Venture Partners. Existing investors, including Ignition Partners, Zetta Venture Partners, Acequia Capital and Plug and Play Ventures, also participated. The new capital is to be spent on accelerating growth, expanding its research and development and entering new markets.

(The Series B also included an additional $5 million in secondary funding, seeing some investors at least partially exit. I understand Tractable’s founders sold a relatively small number of shares as they were permitted to take money off the table. Dalyac declined to comment.)

As we wrap up our call, I note that all of Tractable’s main investors, not including EF, are from the U.S. — something Dalyac says was a deliberate decision after he discovered the gulf between European and U.S. valuations.

“That’s a shame, isn’t it?” I say with my European tech ecosystem hat on.

“It isn’t; it’s enormous exports for the U.K.,” says the Tractable CEO who is French-born but raised in the U.K. “We have, as of today, the vast majority of our headcount in London. The entire product team is in London. The entire R&D team is in London. But most of the revenue comes from the United States. We are making AI an export industry of the U.K.”

Read more: feedproxy.google.com

The post Tractable is applying AI to accident and disaster appraisal appeared first on TheFeedPost.

from WordPress https://ift.tt/goU3rj https://ift.tt/2Aw2lOL via IFTTT

[endenogatai]

Tractable is applying AI to accident and disaster appraisal

“Happy to spend 10 minutes on our vision and the journey we’re on, but then, really, 15 minutes on what we’ve got today, what it is we’ve achieved, what it is our AI does,” says Tractable co-founder and CEO Alexandre Dalyac when I video called him a couple of weeks ago. “You can probably speed up all of that,” I quip back.

The resulting conversation, lasting well over an hour, spanned all of the above and more, including what is required to build a successful AI business and why he and his team think they can help prevent another “AI winter.”

Founded in 2014 by Dalyac, Adrien Cohen and Razvan Ranca after going through company builder Entrepreneur First, London-based Tractable is applying artificial intelligence to accident and disaster recovery. Specifically, through the use of deep learning to automate visual damage appraisal, and therefore help speed up insurance payouts and access to other types of financial aid.

Our AI has already been trained on tens of millions of these cases, so that’s a perfect case of us already having distilled thousands of people’s work experience Alexandre Dalyac

Dalyac launches into what is clearly a well-rehearsed and evidently polished pitch. “We are on a journey to help the world recover faster from accidents and disasters. Our belief is that when accidents and disasters hit, the response could be 10 times faster thanks to AI. So what we mean there is, everything from road accidents, burst piping to large-scale floods and hurricane. Whenever any of these things happen, things get damaged.”

Those things, he says, broadly break down into cars, homes and crops, roughly equating to $1 trillion in damage each year. But, perhaps more importantly, livelihoods get impacted.

“If a car gets damaged, mobility is reduced. If a home gets damaged, shelter is reduced. And if crops get damaged, food is reduced. Across all of those accidents and disasters, we’re talking hundreds of millions of lives affected.”

It is here where a little lateral (and non-artificial) thinking is required. Accident and disaster recovery starts with visual damage appraisal: look at the damage, say how much it’s going to cost, unlock the funds and rebuild. The problem (and Tractable’s opportunity) is that having an appraiser look at a car, house or field can take days to weeks depending on availability — and therefore so can accessing funds to start rebuilding — whereas the claim is that computer vision and AI technology can potentially do the same job in minutes.

“When you assess, that is basically a very powerful but very narrow visual task, which is, look at the damage, how much is it gonna cost? Today, as you can imagine, these kind of assessments are manual. And they take days to weeks. And so you instantly know that with AI that can be 10 times faster,” says Dalyac.

“In some sense this is a perfect class of AI tasks, because it’s very heavy on image classification. And image classification is a task where AI can surpass human performance as of this decade. If you have instant appraisal, that means faster recovery. Hence the mission.”

Dalyac says that part of Tractable’s secret sauce is in the many millions of proprietary labels the company has produced. This has been aided by its patented “interactive machine learning technology,” which allows it to label images faster and cheaper than typical labeling services.

The team’s focus to date has been to train its AI to understand car damage, technology it has already deployed in six countries, seeing the startup work primarily with insurers.

Related to this I’m shown a simple demo of Tractable’s car damage appraisal tool. Dalyac opens a folder of car images on his laptop and uploads them to the software. Within seconds, the AI has seemingly identified the different parts of the car and determined which parts can be repaired and which parts need to be entirely written-off and therefore replaced fully. Each has an AI-generated estimated cost.

It all happens within a matter of minutes, although I have no way of knowing how difficult the pre-determined and fully controlled task is. It’s also unclear how an AI can possibly do the full job of a human assessor based on a limited set of 2D images alone, and without the ability to peek under the hood or undertake further investigations.

“We’re trying to figure out how much damage there is to a vehicle based on photos,” explains Dalyac. “There’s some really tough correlations to pick out, which are: based on the photos of the outside, what’s the internal damage? When you’re a human you are going to have seen and torn down maybe about a thousand to two thousand cars in your whole life of 20 or 30 years of doing that. Our AI has already been trained on tens of millions of these cases, so that’s a perfect case of us already having distilled thousands of people’s work experience. That allows us to get hold of some very challenging correlations that humans just can’t do.”

You need to find real-world use cases that will make a difference, where you can surpass human performance Alexandre Dalyac

With that said, he does concede that a photo doesn’t always contain all of the necessary information, and that it might only have a certain level of accuracy. “You might need to then get a tear-down of the car and get photos of the internal damage. You might even want to get some data from the dashboard. And you can think that as cars get more sensors… the appraisal will be not just visual but also based on IoT data. But that doesn’t detract from the fact that we are convinced that it will be AI that will be doing this entirely.”

What is abundantly clear is Dalyac’s commitment to developing AI technology with real-world use that is commercially viable. If that doesn’t happen, he believes it won’t just be Tractable that will suffer, but the continued belief and investment in AI as a whole. Here, of course, he’s talking about the prospect of another so-called “AI winter,” citing a recent Crunchbase report that says funding for artificial intelligence companies in the U.S. has levelled off and even started to decline at seed stage.

“If you’re trying to make the $15 billion that has been invested into AI not fuck up and lead to something successful that will prevent an AI winter that will lead to continuous improvement, you need a really good return on that asset class. And for that you need those businesses to be successful.

“To make an AI company successful, really successful — not just an acqui-hire, not just an IP exit but a real commercial success that’s going to prevent an AI winter — you need to find real-world use cases that will make a difference, where you can surpass human performance, where you can change the way things work,” he says.

The reference to acqui-hire or IP exit takes on more meaning when you consider that Tractable was in the same cohort at Entrepreneur First as Magic Pony Technology, the AI startup acquired by Twitter for up to $150 million for its image enhancing technology. And most recently, the team behind Bloomsbury AI, another EF company, was acqui-hired by Facebook for $20-30 million.

To ensure that Tractable can continue its mission of applying AI to accident and disaster recovery — and presumably not sell too early — the startup has closed $20 million in Series B investment in a round led by U.S. venture capital firm Insight Venture Partners. Existing investors, including Ignition Partners, Zetta Venture Partners, Acequia Capital and Plug and Play Ventures, also participated. The new capital is to be spent on accelerating growth, expanding its research and development and entering new markets.

(The Series B also included an additional $5 million in secondary funding, seeing some investors at least partially exit. I understand Tractable’s founders sold a relatively small number of shares as they were permitted to take money off the table. Dalyac declined to comment.)

As we wrap up our call, I note that all of Tractable’s main investors, not including EF, are from the U.S. — something Dalyac says was a deliberate decision after he discovered the gulf between European and U.S. valuations.

“That’s a shame, isn’t it?” I say with my European tech ecosystem hat on.

“It isn’t; it’s enormous exports for the U.K.,” says the Tractable CEO who is French-born but raised in the U.K. “We have, as of today, the vast majority of our headcount in London. The entire product team is in London. The entire R&D team is in London. But most of the revenue comes from the United States. We are making AI an export industry of the U.K.”

from RSSMix.com Mix ID 8204425 https://ift.tt/2OwySHf via IFTTT

[thehowtostuff-blog]

“Happy to spend 10 minutes on our vision and the journey we’re on, but then, really, 15 minutes on what we’ve got today, what it is we’ve achieved, what it is our AI does,” says Tractable co-founder and CEO Alexandre Dalyac when I video called him a couple of weeks ago. “You can probably speed up all of that,” I quip back.

The resulting conversation, lasting well over an hour, spanned all of the above and more, including what is required to build a successful AI business and why he and his team think they can help prevent another “AI winter.”

Founded in 2014 by Dalyac, Adrien Cohen and Razvan Ranca after going through company builder Entrepreneur First, London-based Tractable is applying artificial intelligence to accident and disaster recovery. Specifically, through the use of deep learning to automate visual damage appraisal, and therefore help speed up insurance payouts and access to other types of financial aid.

Our AI has already been trained on tens of millions of these cases, so that’s a perfect case of us already having distilled thousands of people’s work experience Alexandre Dalyac

Dalyac launches into what is clearly a well-rehearsed and evidently polished pitch. “We are on a journey to help the world recover faster from accidents and disasters. Our belief is that when accidents and disasters hit, the response could be 10 times faster thanks to AI. So what we mean there is, everything from road accidents, burst piping to large-scale floods and hurricane. Whenever any of these things happen, things get damaged.”

Those things, he says, broadly break down into cars, homes and crops, roughly equating to $1 trillion in damage each year. But, perhaps more importantly, livelihoods get impacted.

“If a car gets damaged, mobility is reduced. If a home gets damaged, shelter is reduced. And if crops get damaged, food is reduced. Across all of those accidents and disasters, we’re talking hundreds of millions of lives affected.”

It is here where a little lateral (and non-artificial) thinking is required. Accident and disaster recovery starts with visual damage appraisal: look at the damage, say how much it’s going to cost, unlock the funds and rebuild. The problem (and Tractable’s opportunity) is that having an appraiser look at a car, house or field can take days to weeks depending on availability — and therefore so can accessing funds to start rebuilding — whereas the claim is that computer vision and AI technology can potentially do the same job in minutes.

“When you assess, that is basically a very powerful but very narrow visual task, which is, look at the damage, how much is it gonna cost? Today, as you can imagine, these kind of assessments are manual. And they take days to weeks. And so you instantly know that with AI that can be 10 times faster,” says Dalyac.

“In some sense this is a perfect class of AI tasks, because it’s very heavy on image classification. And image classification is a task where AI can surpass human performance as of this decade. If you have instant appraisal, that means faster recovery. Hence the mission.”

Dalyac says that part of Tractable’s secret sauce is in the many millions of proprietary labels the company has produced. This has been aided by its patented “interactive machine learning technology,” which allows it to label images faster and cheaper than typical labeling services.

The team’s focus to date has been to train its AI to understand car damage, technology it has already deployed in six countries, seeing the startup work primarily with insurers.

Related to this I’m shown a simple demo of Tractable’s car damage appraisal tool. Dalyac opens a folder of car images on his laptop and uploads them to the software. Within seconds, the AI has seemingly identified the different parts of the car and determined which parts can be repaired and which parts need to be entirely written-off and therefore replaced fully. Each has an AI-generated estimated cost.

It all happens within a matter of minutes, although I have no way of knowing how difficult the pre-determined and fully controlled task is. It’s also unclear how an AI can possibly do the full job of a human assessor based on a limited set of 2D images alone, and without the ability to peek under the hood or undertake further investigations.

“We’re trying to figure out how much damage there is to a vehicle based on photos,” explains Dalyac. “There’s some really tough correlations to pick out, which are: based on the photos of the outside, what’s the internal damage? When you’re a human you are going to have seen and torn down maybe about a thousand to two thousand cars in your whole life of 20 or 30 years of doing that. Our AI has already been trained on tens of millions of these cases, so that’s a perfect case of us already having distilled thousands of people’s work experience. That allows us to get hold of some very challenging correlations that humans just can’t do.”

You need to find real-world use cases that will make a difference, where you can surpass human performance Alexandre Dalyac

With that said, he does concede that a photo doesn’t always contain all of the necessary information, and that it might only have a certain level of accuracy. “You might need to then get a tear-down of the car and get photos of the internal damage. You might even want to get some data from the dashboard. And you can think that as cars get more sensors… the appraisal will be not just visual but also based on IoT data. But that doesn’t detract from the fact that we are convinced that it will be AI that will be doing this entirely.”

What is abundantly clear is Dalyac’s commitment to developing AI technology with real-world use that is commercially viable. If that doesn’t happen, he believes it won’t just be Tractable that will suffer, but the continued belief and investment in AI as a whole. Here, of course, he’s talking about the prospect of another so-called “AI winter,” citing a recent Crunchbase report that says funding for artificial intelligence companies in the U.S. has levelled off and even started to decline at seed stage.

“If you’re trying to make the $15 billion that has been invested into AI not fuck up and lead to something successful that will prevent an AI winter that will lead to continuous improvement, you need a really good return on that asset class. And for that you need those businesses to be successful.

“To make an AI company successful, really successful — not just an acqui-hire, not just an IP exit but a real commercial success that’s going to prevent an AI winter — you need to find real-world use cases that will make a difference, where you can surpass human performance, where you can change the way things work,” he says.

The reference to acqui-hire or IP exit takes on more meaning when you consider that Tractable was in the same cohort at Entrepreneur First as Magic Pony Technology, the AI startup acquired by Twitter for up to $150 million for its image enhancing technology. And most recently, the team behind Bloomsbury AI, another EF company, was acqui-hired by Facebook for $20-30 million.

To ensure that Tractable can continue its mission of applying AI to accident and disaster recovery — and presumably not sell too early — the startup has closed $20 million in Series B investment in a round led by U.S. venture capital firm Insight Venture Partners. Existing investors, including Ignition Partners, Zetta Venture Partners, Acequia Capital and Plug and Play Ventures, also participated. The new capital is to be spent on accelerating growth, expanding its research and development and entering new markets.

(The Series B also included an additional $5 million in secondary funding, seeing some investors at least partially exit. I understand Tractable’s founders sold a relatively small number of shares as they were permitted to take money off the table. Dalyac declined to comment.)

As we wrap up our call, I note that all of Tractable’s main investors, not including EF, are from the U.S. — something Dalyac says was a deliberate decision after he discovered the gulf between European and U.S. valuations.

“That’s a shame, isn’t it?” I say with my European tech ecosystem hat on.

“It isn’t; it’s enormous exports for the U.K.,” says the Tractable CEO who is French-born but raised in the U.K. “We have, as of today, the vast majority of our headcount in London. The entire product team is in London. The entire R&D team is in London. But most of the revenue comes from the United States. We are making AI an export industry of the U.K.”

from TechCrunch https://ift.tt/2OwySHf

[fmservers]

Tractable is applying AI to accident and disaster appraisal

“Happy to spend 10 minutes on our vision and the journey we’re on, but then, really, 15 minutes on what we’ve got today, what it is we’ve achieved, what it is our AI does,” says Tractable co-founder and CEO Alexandre Dalyac when I video called him a couple of weeks ago. “You can probably speed up all of that,” I quip back.

The resulting conversation, lasting well over an hour, spanned all of the above and more, including what is required to build a successful AI business and why he and his team think they can help prevent another “AI winter.”

Founded in 2014 by Dalyac, Adrien Cohen and Razvan Ranca after going through company builder Entrepreneur First, London-based Tractable is applying artificial intelligence to accident and disaster recovery. Specifically, through the use of deep learning to automate visual damage appraisal, and therefore help speed up insurance payouts and access to other types of financial aid.

Our AI has already been trained on tens of millions of these cases, so that’s a perfect case of us already having distilled thousands of people’s work experience Alexandre Dalyac

Dalyac launches into what is clearly a well-rehearsed and evidently polished pitch. “We are on a journey to help the world recover faster from accidents and disasters. Our belief is that when accidents and disasters hit, the response could be 10 times faster thanks to AI. So what we mean there is, everything from road accidents, burst piping to large-scale floods and hurricane. Whenever any of these things happen, things get damaged.”

Those things, he says, broadly break down into cars, homes and crops, roughly equating to $1 trillion in damage each year. But, perhaps more importantly, livelihoods get impacted.

“If a car gets damaged, mobility is reduced. If a home gets damaged, shelter is reduced. And if crops get damaged, food is reduced. Across all of those accidents and disasters, we’re talking hundreds of millions of lives affected.”

It is here where a little lateral (and non-artificial) thinking is required. Accident and disaster recovery starts with visual damage appraisal: look at the damage, say how much it’s going to cost, unlock the funds and rebuild. The problem (and Tractable’s opportunity) is that having an appraiser look at a car, house or field can take days to weeks depending on availability — and therefore so can accessing funds to start rebuilding — whereas the claim is that computer vision and AI technology can potentially do the same job in minutes.

“When you assess, that is basically a very powerful but very narrow visual task, which is, look at the damage, how much is it gonna cost? Today, as you can imagine, these kind of assessments are manual. And they take days to weeks. And so you instantly know that with AI that can be 10 times faster,” says Dalyac.

“In some sense this is a perfect class of AI tasks, because it’s very heavy on image classification. And image classification is a task where AI can surpass human performance as of this decade. If you have instant appraisal, that means faster recovery. Hence the mission.”

Dalyac says that part of Tractable’s secret sauce is in the many millions of proprietary labels the company has produced. This has been aided by its patented “interactive machine learning technology,” which allows it to label images faster and cheaper than typical labeling services.

The team’s focus to date has been to train its AI to understand car damage, technology it has already deployed in six countries, seeing the startup work primarily with insurers.

Related to this I’m shown a simple demo of Tractable’s car damage appraisal tool. Dalyac opens a folder of car images on his laptop and uploads them to the software. Within seconds, the AI has seemingly identified the different parts of the car and determined which parts can be repaired and which parts need to be entirely written-off and therefore replaced fully. Each has an AI-generated estimated cost.

It all happens within a matter of minutes, although I have no way of knowing how difficult the pre-determined and fully controlled task is. It’s also unclear how an AI can possibly do the full job of a human assessor based on a limited set of 2D images alone, and without the ability to peek under the hood or undertake further investigations.

“We’re trying to figure out how much damage there is to a vehicle based on photos,” explains Dalyac. “There’s some really tough correlations to pick out, which are: based on the photos of the outside, what’s the internal damage? When you’re a human you are going to have seen and torn down maybe about a thousand to two thousand cars in your whole life of 20 or 30 years of doing that. Our AI has already been trained on tens of millions of these cases, so that’s a perfect case of us already having distilled thousands of people’s work experience. That allows us to get hold of some very challenging correlations that humans just can’t do.”

You need to find real-world use cases that will make a difference, where you can surpass human performance Alexandre Dalyac

With that said, he does concede that a photo doesn’t always contain all of the necessary information, and that it might only have a certain level of accuracy. “You might need to then get a tear-down of the car and get photos of the internal damage. You might even want to get some data from the dashboard. And you can think that as cars get more sensors… the appraisal will be not just visual but also based on IoT data. But that doesn’t detract from the fact that we are convinced that it will be AI that will be doing this entirely.”

What is abundantly clear is Dalyac’s commitment to developing AI technology with real-world use that is commercially viable. If that doesn’t happen, he believes it won’t just be Tractable that will suffer, but the continued belief and investment in AI as a whole. Here, of course, he’s talking about the prospect of another so-called “AI winter,” citing a recent Crunchbase report that says funding for artificial intelligence companies in the U.S. has levelled off and even started to decline at seed stage.

“If you’re trying to make the $15 billion that has been invested into AI not fuck up and lead to something successful that will prevent an AI winter that will lead to continuous improvement, you need a really good return on that asset class. And for that you need those businesses to be successful.

“To make an AI company successful, really successful — not just an acqui-hire, not just an IP exit but a real commercial success that’s going to prevent an AI winter — you need to find real-world use cases that will make a difference, where you can surpass human performance, where you can change the way things work,” he says.

The reference to acqui-hire or IP exit takes on more meaning when you consider that Tractable was in the same cohort at Entrepreneur First as Magic Pony Technology, the AI startup acquired by Twitter for up to $150 million for its image enhancing technology. And most recently, the team behind Bloomsbury AI, another EF company, was acqui-hired by Facebook for $20-30 million.

To ensure that Tractable can continue its mission of applying AI to accident and disaster recovery — and presumably not sell too early — the startup has closed $20 million in Series B investment in a round led by U.S. venture capital firm Insight Venture Partners. Existing investors, including Ignition Partners, Zetta Venture Partners, Acequia Capital and Plug and Play Ventures, also participated. The new capital is to be spent on accelerating growth, expanding its research and development and entering new markets.

(The Series B also included an additional $5 million in secondary funding, seeing some investors at least partially exit. I understand Tractable’s founders sold a relatively small number of shares as they were permitted to take money off the table. Dalyac declined to comment.)

As we wrap up our call, I note that all of Tractable’s main investors, not including EF, are from the U.S. — something Dalyac says was a deliberate decision after he discovered the gulf between European and U.S. valuations.

“That’s a shame, isn’t it?” I say with my European tech ecosystem hat on.

“It isn’t; it’s enormous exports for the U.K.,” says the Tractable CEO who is French-born but raised in the U.K. “We have, as of today, the vast majority of our headcount in London. The entire product team is in London. The entire R&D team is in London. But most of the revenue comes from the United States. We are making AI an export industry of the U.K.”

Via Steve O'Hear https://techcrunch.com

[clusterassets]

New world news from Time: Our ‘Us Vs Them’ World: 5 Reasons Why Globalism Is Failing

This world is an increasingly polarized place. Divisions are growing not just between countries but within them. In my book Us vs Them: The Failure of Globalism I set out why this is happening now. Here, five key factors dividing today’s world into “us” and “them.”

1. Economics

The growing division of wealth between haves and have-nots has received plenty of attention in recent years—when 42 people in the world have the same wealth as the bottom 50 percent, the headlines write themselves. But attention is one thing, and action another.

Free trade remains the best way we know to spur sustainable economic growth for the world as a whole. Moving goods, services, investment, and innovation as efficiently as possible generates consumption, a virtuous cycle that has seen the world economy soar nearly 700 percent since 1980. Hundreds of millions around the world have escaped poverty, largely thanks to globalization.

But not all countries and peoples share in this virtuous cycle equally. Countries and specific sets of workers lose when jobs and opportunities are sent abroad for the sake of profit margins. And when infrastructure spending, public school systems, health care and the like are tied directly to the economic fortunes of a community, it compounds problems of inequality. It’s taken a couple of generations—and the sharp shock of Brexit plus the election of Donald Trump—for this to become unavoidably obvious for globalization’s still-complacent winners in Europe and the U.S.

2. Society and culture

In a globalized world, people flow across borders too. When workers see threats to their lives, livelihoods, status, and entitlements, they demand walls — barriers against cheap labor and unfamiliar faces—or what you might call “them.”

Donald Trump understood this better than any of his political rivals in the U.S. Through storm after storm, and despite his inability to “drain the swamp” in Washington, his most loyal supporters stick with him, because no one else in the United States can credibly promise to defend their interests against establishment disdain.

Most politicians call for unity. Trump speaks of “us vs them” and continues to reap the rewards.

3. Security

Cultural divides also bleed into the security realm. Global trade demands geopolitical stability. Stability requires leaders willing to do more so others can do less, who use their power to impose the compromises on which multinational progress depends.

But Trump defeated 16 Republicans and Hillary Clinton with promises that America would defend no interests but her own. He derided endless pointless wars and the presidents responsible for them. And while it was America’s political and military establishment that argued for the necessity of all these wars, they were wars waged on the backs of working-class American men and women. When they returned home, they found a world not much safer than the ones they risked their lives to defend. Even worse, instead of being treated as heroes, they were barely treated at all by a dysfunctional veterans administration system.

And then there’s the bundling of immigration concerns with terrorist fears. In the U.S., it’s a narrative that has been politicized for years but has little basis in reality. Europe is another matter; some European communities have historically had a more difficult time integrating immigrant populations, and the threat posed by homegrown jihadi militants, radicalized in slums like the banlieues of Paris and neighborhoods like Molenbeek in Belgium, is all too real.

4. Technology and filter bubbles

The Internet has long been used to connect folks seeking like-minded individuals in chat forums and online communities. But with the advent and explosion of social media, echo chambers took on a life of their own. Today, it takes work to find someone with whom you fundamentally disagree—people gravitate to others who share their values and assumptions about their communities and the world.

Human nature isn’t only to blame, though. Tech and media companies now grow their bottom lines by maximizing the amount of time you spend engaging with content on their platforms. Algorithms are designed to show you content you’re expected to “like” and engage with, with the result being narrower demographic groups that help advertisement targeting and data collection. The result is ever more political fragmentation in the cyber sphere, which manifests itself more profoundly in the real world with each election cycle.

5. Technology and automation

And the tech revolution is still only in its infancy. People talk about the coming era of automation and artificial intelligence (AI) as the “fourth industrial revolution,” placing it (somewhat comfortingly) in historical context. But we’ve never seen anything on the scale of what’s to come—automation and AI are expected to cost 400 to 800 million people their jobs by 2030. T

The world may produce more with robots at the helm, but the economic gains will go mostly to the few who control the technology; hundreds of millions of others will be left with less work to do (if they find work at all). And for all the talk of retraining to prepare people for this automated future, few of those plans have come to fruition. If anything, we should be preparing for a “post-industrial revolution,” one that looks set to widen the chasm between “us” and “them” still further.

And to think that all this fracturing into groups, pitting us against them, is happening at a time when the global economy is growing at a solid clip. In today’s polarized environment, one global economic stumble may be enough to shatter our interconnected world completely.

May 04, 2018 at 05:56PM ClusterAssets Inc., https://ClusterAssets.wordpress.com

[omnipop-mag-blog]

http://time.com/5264653/us-vs-them-globalism-failing/

This world is an increasingly polarized place. Divisions are growing not just between countries but within them. In my book Us vs Them: The Failure of Globalism I set out why this is happening now. Here, five key factors dividing today’s world into “us” and “them.”

1. Economics

The growing division of wealth between haves and have-nots has received plenty of attention in recent years—when 42 people in the world have the same wealth as the bottom 50 percent, the headlines write themselves. But attention is one thing, and action another.

Free trade remains the best way we know to spur sustainable economic growth for the world as a whole. Moving goods, services, investment, and innovation as efficiently as possible generates consumption, a virtuous cycle that has seen the world economy soar nearly 700 percent since 1980. Hundreds of millions around the world have escaped poverty, largely thanks to globalization.

But not all countries and peoples share in this virtuous cycle equally. Countries and specific sets of workers lose when jobs and opportunities are sent abroad for the sake of profit margins. And when infrastructure spending, public school systems, health care and the like are tied directly to the economic fortunes of a community, it compounds problems of inequality. It’s taken a couple of generations—and the sharp shock of Brexit plus the election of Donald Trump—for this to become unavoidably obvious for globalization’s still-complacent winners in Europe and the U.S.

2. Society and culture

In a globalized world, people flow across borders too. When workers see threats to their lives, livelihoods, status, and entitlements, they demand walls — barriers against cheap labor and unfamiliar faces—or what you might call “them.”

Donald Trump understood this better than any of his political rivals in the U.S. Through storm after storm, and despite his inability to “drain the swamp” in Washington, his most loyal supporters stick with him, because no one else in the United States can credibly promise to defend their interests against establishment disdain.

Most politicians call for unity. Trump speaks of “us vs them” and continues to reap the rewards.

3. Security

Cultural divides also bleed into the security realm. Global trade demands geopolitical stability. Stability requires leaders willing to do more so others can do less, who use their power to impose the compromises on which multinational progress depends.

But Trump defeated 16 Republicans and Hillary Clinton with promises that America would defend no interests but her own. He derided endless pointless wars and the presidents responsible for them. And while it was America’s political and military establishment that argued for the necessity of all these wars, they were wars waged on the backs of working-class American men and women. When they returned home, they found a world not much safer than the ones they risked their lives to defend. Even worse, instead of being treated as heroes, they were barely treated at all by a dysfunctional veterans administration system.

And then there’s the bundling of immigration concerns with terrorist fears. In the U.S., it’s a narrative that has been politicized for years but has little basis in reality. Europe is another matter; some European communities have historically had a more difficult time integrating immigrant populations, and the threat posed by homegrown jihadi militants, radicalized in slums like the banlieues of Paris and neighborhoods like Molenbeek in Belgium, is all too real.

4. Technology and filter bubbles

The Internet has long been used to connect folks seeking like-minded individuals in chat forums and online communities. But with the advent and explosion of social media, echo chambers took on a life of their own. Today, it takes work to find someone with whom you fundamentally disagree—people gravitate to others who share their values and assumptions about their communities and the world.

Human nature isn’t only to blame, though. Tech and media companies now grow their bottom lines by maximizing the amount of time you spend engaging with content on their platforms. Algorithms are designed to show you content you’re expected to “like” and engage with, with the result being narrower demographic groups that help advertisement targeting and data collection. The result is ever more political fragmentation in the cyber sphere, which manifests itself more profoundly in the real world with each election cycle.

5. Technology and automation

And the tech revolution is still only in its infancy. People talk about the coming era of automation and artificial intelligence (AI) as the “fourth industrial revolution,” placing it (somewhat comfortingly) in historical context. But we’ve never seen anything on the scale of what’s to come—automation and AI are expected to cost 400 to 800 million people their jobs by 2030. T

The world may produce more with robots at the helm, but the economic gains will go mostly to the few who control the technology; hundreds of millions of others will be left with less work to do (if they find work at all). And for all the talk of retraining to prepare people for this automated future, few of those plans have come to fruition. If anything, we should be preparing for a “post-industrial revolution,” one that looks set to widen the chasm between “us” and “them” still further.

And to think that all this fracturing into groups, pitting us against them, is happening at a time when the global economy is growing at a solid clip. In today’s polarized environment, one global economic stumble may be enough to shatter our interconnected world completely.

The post New world news from Time: Our ‘Us Vs Them’ World: 5 Reasons Why Globalism Is Failing appeared first on OMNI POP MAG.

[nikolasfuturist]

Each week on a Wednesday Nikolas Badminton, Futurist highlights the top stories from the past week relating to the incredible rise of artificial intelligence and its application in society, the workplace, in cities, and in our lives.

In Artificial Intelligence Bulletin – AI Will Put You in Prison we look at how algorithms can be used for incarceration, teaching cars to drive, the reality behind the Luddite’s cause, democratizing AI, and putting 10,000 people to work.

Sent to Prison by a Software Program’s Secret Algorithms

When Chief Justice John G. Roberts Jr. visited Rensselaer Polytechnic Institute last month, he was asked a startling question, one with overtones of science fiction.

“Can you foresee a day,” asked Shirley Ann Jackson, president of the college in upstate New York, “when smart machines, driven with artificial intelligences, will assist with courtroom fact-finding or, more controversially even, judicial decision-making?”

The chief justice’s answer was more surprising than the question. “It’s a day that’s here,” he said, “and it’s putting a significant strain on how the judiciary goes about doing things.”

He may have been thinking about the case of a Wisconsin man, Eric L. Loomis, who was sentenced to six years in prison based in part on a private company’s proprietary software. Mr. Loomis says his right to due process was violated by a judge’s consideration of a report generated by the software’s secret algorithm, one Mr. Loomis was unable to inspect or challenge.

Read more at New York Times

Open Source Stories: Road to AI

How do you teach a car to drive? For many self-driving car makers and artificial intelligence researchers, the answer starts with data and sharing.

Luddites have been getting a bad rap for 200 years. But, turns out, they were right

Things did not end well for the Luddites. The group of weavers and textile artisans in early 1800s were crushed by the British government after resisting the destruction of their livelihoods by industrialization. History, in one of its callous twists, recast their story from a workers’ revolt for fair treatment to a short-sighted war against technology and progress.

The truth is that the Luddites were the skilled, middle-class workers of their time. After centuries on more-or-less good terms with merchants who sold their goods, their lives were upended by machines replacing them with low-skilled, low-wage laborers in dismal factories. To ease the transition, the Luddites sought to negotiate conditions similar to those underlying capitalist democracies today: taxes to fund workers’ pensions, a minimum wage, and adherence to minimum labor standards.

Those bargaining attempts were rebuffed by most factory owners. The Luddites then began months of “machine breaking” in 1811-1812, smashing the weaving frames, in a last ditch effort to bring their new bosses to the table. At the behest of factory owners, the British Parliament declared machine breaking a capital offense and sent 14,000 troops to the English countryside to put down the uprising. Dozens of Luddites were executed or exiled to Australia. The crushed rebellion cleared the way for horrific working conditions of the Industrial Revolution yet to come.

Read more at Quartz

Democratizing Artificial Intelligence

Artificial Intelligence is the next technological frontier, and it has the potential to make or break the world order. The AI revolution could pull the “bottom billion” out of poverty and transform dysfunctional institutions, or it could entrench injustice and increase inequality. The outcome will depend on how we manage the coming changes.

Unfortunately, when it comes to managing technological revolutions, humanity has a rather poor track record. Consider the Internet, which has had an enormous impact on societies worldwide, changing how we communicate, work, and occupy ourselves. And it has disrupted some economic sectors, forced changes to long-established business models, and created a few entirely new industries.

But the Internet has not brought the kind of comprehensive transformation that many anticipated. It certainly didn’t resolve the big problems, such as eradicating poverty or enabling us to reach Mars. As PayPal co-founder Peter Thiel once noted: “We wanted flying cars; instead, we got 140 characters.”

In fact, in some ways, the Internet has exacerbated our problems. While it has created opportunities for ordinary people, it has created even more opportunities for the wealthiest and most powerful. A recent study by researchers at the LSE reveals that the Internet has increased inequality, with educated, high-income people deriving the greatest benefits online and multinational corporations able to grow massively – while evading accountability.

Read more at Project Syndicate

Automation Jobs Will Put 10,000 Humans to Work, Study Says

It’s going to take a lot of humans to create the kind of artificial intelligence that could replace truckers, financial analysts, and customer service representatives with robots. U.S. employers will spend more than $650 million on annual salaries for 10,000 jobs in AI this year, according to a study from career and hiring data firm Paysa.

The 2-year-old firm touts itself as the only platform to use AI to deliver personalized job and salary recommendations. It was founded by Chris Bolte, Zachary Poley, Nikhil Raj and Patrick Harrington — all formerly of Walmart Labs and Walmart’s engineering and product teams.

The firm uses millions of data points like job openings, resumes, and compensation to determine the market value of individual skills. Once a worker creates a profile on the site, they can look at how their salary compares to others with the same job title and experience.

Read more at FORTUNE

The post Artificial Intelligence Bulletin – AI Will Put You in Prison appeared first on Nikolas Badminton, Futurist Speaker.

[itviconsultants]

Should software developers have a code of ethics?

Should software developers have a code of ethics?: First, do no harm. This is the underlying message of the Hippocratic Oath historically taken by physicians to show they will abide by an ethical code of conduct. Plumbers, construction workers, law enforcement -- almost any professional whose work impacts the public must abide by some sort of ethical code of conduct. There's one fairly notable exception: technology. While there are organization-- and company-specific codes of conduct -- like these guidelines from the Association for Computing Machinery (ACM) and the Institute of Electrical and Electronics Engineers - Computer Science (IEEE-CS) joint task force on software engineering ethics professional practices, there's no one all-encompassing set of standards that includes the entire industry. But maybe there should be. In 2015, independent tests revealed that Volkswagen engineers programmed cars to cheat emissions standards. In the wake of the 2016 U.S. presidential election, Facebook -- among others -- is grappling with an epidemic of fake news. The nation is struggling to come to grips with alleged Russian hacking and interference in our elections. And the current president-elect campaigned on the promise to build on the existing (or build a new) Muslim registry to track members of that faith. [ Related story: Workplace bullying a costly epidemic in the enterprise ] Good versus evil? These are just a few examples of how software can be used for nefarious purposes; there's no way to know definitively every possible outcome of the development and use of every piece of technology, every line of code. So, it's up to those who design and build the products, software packages, the apps and solutions that we use daily to do the right thing. That's a lot of pressure. It's also difficult to navigate what's right and wrong if you're pressured to meet deadlines, or your livelihood's on the line; a code of ethics can provide context and framework for professionals to fall back on, says Dave West, product owner at Scrum.org. And while he'd like to see such a thing, he says it's understandable that such a diverse-thinking group might not be able to agree on all aspects of what such a code would entail. "I would love to see a standardized, industry code of ethics. We do have our own that falls under our mission of improving the profession of crafting software… And we feel like that is a solid foundation for anyone to fall back on if they are feeling uncertain about any part of their job responsibilities, because they can step back and look at those values and say, 'Am I doing the right thing, here, based on these things I believe in?'" West says. The debate about ethics in software development has raged on for as long as the profession's been around. It can be nearly impossible to assess all the potential applications of a technology, good and bad, and that's both the beauty and the horror of the issue, says Shon Burton, founder and CEO, HiringSolved, which uses AI to help companies identify diverse talent. [ Related story: Why your business can't afford toxic employees ] Any tool can be a weapon "Any tool can be a weapon depending on how you use it. For us, using AI and automation -- the stuff I can think about now that we're close to it, I can see good and bad, and both are easily accessible. For our applications, we can help clients screen for diverse candidates. But we see, also, that it could be used to screen out people with different ethnicities, races, gender. We have a code of conduct internally that we all adhere to. But we understand the potential and the unintended consequences," Burton says. In the absence of an industry-wide set of ethical standards, individuals and even some corporate entities are making public stands behind their values. In December, the NeverAgain.tech movement circulated a pledge to resist "…build[ing] a database of people based on their Constitutionally-protected religious beliefs. We refuse to facilitate mass deportations of people the government believes to be undesirable," the pledge reads. It's now gathered more than 2,500 signatures. GrubHub CEO Matt Maloney took a lot of heat for his stand against hateful, demeaning and discriminatory actions and language. And Oracle executive George Polisner very publicly resigned his position in response to his former employer's co-CEO accepting a role in the incoming presidential administration. It can be difficult to know where the line exists between right and wrong in this context, even if you're walking it. While one standardized code of ethics could be a solution, it may be more important to teach people how to ask the right questions, says Scrum.org's West. "Personally, I'd love to see more education on teaching ethics than is presently available, especially in a professional context rather than just a course about theory, because ethics in isolation won't work unless it's part of a broader professional standards. There's also an issue, though, of the fact that often, individuals can't build software alone, and they're also not making these 'wrong' decisions all at once, but incrementally," West says. [ Related story: What makes a company a 'best place to work'? ] Teaching people to ask the right questions involves understanding what the questions are, says Burton, and that everyone's values are different; some individuals have no problem working on software that runs nuclear reactors, or developing targeting systems for drones, or smart bombs, or military craft. "The truth is, we've been here before, and we're already making strides toward mitigating risks and unintended consequences. It's not even a question of can we build it anymore, because we know the technology and capability is out there to build whatever we can think of. The questions should be around should it be built, what are the fail-safes, and what can we do to make sure we're having the least harmful impact we can?" he says. There's no one "right answer" here, and a code of ethics certainly won't put all the ethical issues to rest. But it could be a good place to start if individuals and organizations want to harness the great power of technology to create solutions that serve the greater good. Related Video To read this article in full or to leave a comment, please click here http://dlvr.it/N4ymRG #CIO #ITStrategy

[brianlichtig]

Should software developers have a code of ethics?

First, do no harm. This is the underlying message of the Hippocratic Oath historically taken by physicians to show they will abide by an ethical code of conduct. Plumbers, construction workers, law enforcement -- almost any professional whose work impacts the public must abide by some sort of ethical code of conduct.

There's one fairly notable exception: technology. While there are organization-- and company-specific codes of conduct -- like these guidelines from the Association for Computing Machinery (ACM) and the Institute of Electrical and Electronics Engineers - Computer Science (IEEE-CS) joint task force on software engineering ethics professional practices, there's no one all-encompassing set of standards that includes the entire industry.

But maybe there should be. In 2015, independent tests revealed that Volkswagen engineers programmed cars to cheat emissions standards. In the wake of the 2016 U.S. presidential election, Facebook -- among others -- is grappling with an epidemic of fake news. The nation is struggling to come to grips with alleged Russian hacking and interference in our elections. And the current president-elect campaigned on the promise to build on the existing (or build a new) Muslim registry to track members of that faith.

[ Related story: Workplace bullying a costly epidemic in the enterprise ]

Good versus evil?

These are just a few examples of how software can be used for nefarious purposes; there's no way to know definitively every possible outcome of the development and use of every piece of technology, every line of code. So, it's up to those who design and build the products, software packages, the apps and solutions that we use daily to do the right thing. That's a lot of pressure.

It's also difficult to navigate what's right and wrong if you're pressured to meet deadlines, or your livelihood's on the line; a code of ethics can provide context and framework for professionals to fall back on, says Dave West, product owner at Scrum.org. And while he'd like to see such a thing, he says it's understandable that such a diverse-thinking group might not be able to agree on all aspects of what such a code would entail.

"I would love to see a standardized, industry code of ethics. We do have our own that falls under our mission of improving the profession of crafting software… And we feel like that is a solid foundation for anyone to fall back on if they are feeling uncertain about any part of their job responsibilities, because they can step back and look at those values and say, 'Am I doing the right thing, here, based on these things I believe in?'" West says.

The debate about ethics in software development has raged on for as long as the profession's been around. It can be nearly impossible to assess all the potential applications of a technology, good and bad, and that's both the beauty and the horror of the issue, says Shon Burton, founder and CEO, HiringSolved, which uses AI to help companies identify diverse talent.

[ Related story: Why your business can't afford toxic employees ]

Any tool can be a weapon

"Any tool can be a weapon depending on how you use it. For us, using AI and automation -- the stuff I can think about now that we're close to it, I can see good and bad, and both are easily accessible. For our applications, we can help clients screen for diverse candidates. But we see, also, that it could be used to screen out people with different ethnicities, races, gender. We have a code of conduct internally that we all adhere to. But we understand the potential and the unintended consequences," Burton says.

In the absence of an industry-wide set of ethical standards, individuals and even some corporate entities are making public stands behind their values. In December, the NeverAgain.tech movement circulated a pledge to resist "…build[ing] a database of people based on their Constitutionally-protected religious beliefs. We refuse to facilitate mass deportations of people the government believes to be undesirable," the pledge reads. It's now gathered more than 2,500 signatures.

GrubHub CEO Matt Maloney took a lot of heat for his stand against hateful, demeaning and discriminatory actions and language. And Oracle executive George Polisner very publicly resigned his position in response to his former employer's co-CEO accepting a role in the incoming presidential administration.

It can be difficult to know where the line exists between right and wrong in this context, even if you're walking it. While one standardized code of ethics could be a solution, it may be more important to teach people how to ask the right questions, says Scrum.org's West.

"Personally, I'd love to see more education on teaching ethics than is presently available, especially in a professional context rather than just a course about theory, because ethics in isolation won't work unless it's part of a broader professional standards. There's also an issue, though, of the fact that often, individuals can't build software alone, and they're also not making these 'wrong' decisions all at once, but incrementally," West says.

[ Related story: What makes a company a 'best place to work'? ]

Teaching people to ask the right questions involves understanding what the questions are, says Burton, and that everyone's values are different; some individuals have no problem working on software that runs nuclear reactors, or developing targeting systems for drones, or smart bombs, or military craft.

"The truth is, we've been here before, and we're already making strides toward mitigating risks and unintended consequences. It's not even a question of can we build it anymore, because we know the technology and capability is out there to build whatever we can think of. The questions should be around should it be built, what are the fail-safes, and what can we do to make sure we're having the least harmful impact we can?" he says.

There's no one "right answer" here, and a code of ethics certainly won't put all the ethical issues to rest. But it could be a good place to start if individuals and organizations want to harness the great power of technology to create solutions that serve the greater good.

Related Video

To read this article in full or to leave a comment, please click here

from CIO http://www.cio.com/article/3156565/developer/should-software-developers-have-a-code-of-ethics.html#tk.rss_all Baltimore IT Support