#able to take that story and that autonomy without the threat or forced input of colonialism

[cumbiazevran]

one of the things i often think about in terms of how Elf narratives are written in the game is how little BioWare devs understand colonisation and it’s related struggles.

it’s from the constant framing of the Dalish as not really knowing what they lost, to Solas’ plans. don’t get me wrong, i love the bastard egg as a narrative device, and i’m not immune to posing him as a foil in my Lavellan’s inquisitors own narrative. however, the only people i’ve ever come in contact with that are desperate to reclaim imperial pasts are white people, not people of colour.

because no matter how much they whitewash elves in the game, the truth is theirs is a “narrative of colour”, so to speak. it’s a narrative of marginalisation and peoples who have seen outside of Whiteness through one device or the other. it’s about the layers and layers of colonisation poc have to deal with every day.

if you’re going to put that into an entire franchise, you should do it responsibly, though, granted, this is BioWare we’re talking about.

but i digress: my point here is i never stop being baffled about how through ignorance or willingness (it makes little difference in the outcome) the continue to write Elvish narrative in the games as if they didn’t know what they lost, and their only way through is Restoring An Empire. this isn’t about unreliable narrators or plot devices, it’s about the awareness that does exist that events like colonialism cannot be undone. they are wounds that take different forms depending on where you stand, but they never close, not really. they change history and that’s that: there’s no going back. ever.

and people know this. the idea of restoring empires is so tied to whiteness to me, and so drastically different to anti-colonialism. it’s so telling in the approach of the devs.

[north-of-annwn]

Ok so I'm not going to do this anonymously because i don't fear getting chastised for my own ignorance but there are other alternatives to abortion aren't there? I mean i honestly don't understand this bill nonsense but it mostly at least to me sounds like it's just to keep children alive. I mean there are adoption centers and people who will actually pay women who are pregnant to act as surrogates. Why is anti abortion so bad? And how is this a woman's autonomy probpem. Please educate me

First, I want to thank you for acknowledging that your perspective on this may be informed from a place of systemic oppression of AFAB people, and for seeking out information. When people add on to this post with the purpose to educate, I implore you all to remember this person is seeking information. Please avoid shaming them or ridiculing them.

Let’s first address your questions:

1) “There are other alternatives to abortion aren’t there?”

Let’s first define abortion. “In medicine, an abortion is the premature exit of the products of conception (the fetus, fetal membranes, and placenta) from the uterus. It is the loss of a pregnancy and does not refer to why that pregnancy was lost. A spontaneous abortion is the same as a miscarriage. The miscarriage of three or more consecutive pregnancies is termed habitual abortion or recurrent pregnancy loss”  (Shiel MD, MedicineNet).

¼ womxn will have abortions in their lifetime. Abortion is a medical procedure that can be requested or required for a lot of different reasons:

The pregnant person may not be able to carry an embryo to term safely.

The pregnant person may not have the financial support to pay for the medical bills that pregnancy costs in the US (prenatal and delivery alone can cost around $18k).  

I also want to add that people in this country are not given any kind of financial support for the time taken off for prenatal or postnatal care. Being out of work for this time could mean entering extreme poverty.

The pregnant person may not have the financial support or stability of lifestyle to support a child.

The pregnant person may not be physically up to the task of carrying a child to term and delivering. Not all womb-having people are up to what childbirth does to the body. Childbirth is one of the most dangerous things that a body can be put through.  In the US we’re just under 20 maternal deaths per 100,000 births, which is the highest in the developed world. Some undeveloped countries have better stats than we do

Abortion may be required as an emergency life-saving procedure for the pregnant person. And waiting for approval by a committee could mean the death of that person.

Medical interference can also be needed if the embryo has already been determined unviable (basically will not ever have life) because having dead tissue remain in the womb will kill the person. Wombs don’t always do what they’re supposed to and often they will still act as if the pregnancy is going along normally when the embryo stopped growing and forming.

Abortion as a medical procedure is part of basic reproductive healthcare. Denying it is like denying the use of a c-section or blood transfusions.

I also want to add that many of these GOP states are seeking to classify any and all contraceptives as “abortion” as well. This isn’t included in this bill specifically but it’s been named as part of their agendas.

2) “I mean I honestly don’t understand this bill nonsense but it mostly at least to me sounds like it’s just to keep children alive.”According to the CDC, 91.1% of abortions are performed at ≤13 weeks’ gestation. At this time, this is an embryo and fetal tissue. It’s not a child. Pro-life people are placing the eventual *possible* life of a being that isn’t even formed yet above the autonomy and rights of a living human being (the pregnant person). A zygote without a brain or the ability to survive outside the womb is not a person, and therefore not a child. We have determined that something without brain activity is not alive. People with wombs are not incubators. This is not the sum of our existence.

Right now you cannot force a person to give blood or organs in life-saving situations. Why should it be okay to force a person to donate their entire body as an incubator if they don’t want to, which has health complications, and long-lasting effects on the body? We even afford humans that are DEAD more rights than womb-having people in this country. It is illegal to take organs or tissue from dead bodies with no brain activity without consent, but it’s legal to force a living person to act as an incubator for tissue and chromosomes that aren’t even formed to make a person yet?

Also, this bill has SO much more nuanced support for the oppression of women than just keeping “children” alive. This affords the state the right to investigate any suspicion of “intentional abortion.” This means, if a person miscarries, they may be subject to invasive investigation and murder charges on top of grieving for their loss and recovering medically. This bill also in no certain terms basically considers all womb-having people in their state to be the property of the state by allowing people to be extradited and charged if they have a LEGAL abortion procedure in another state.

3) “I mean there are adoption centers and people who will actually pay women who are pregnant to act as surrogates. Why is anti-abortion so bad?” We currently have 108,000 foster children up for adoption right this second in the US. This doesn’t even include unwanted pregnancies being given to private adoption agencies. Adopt one if you want to save a child, but forcing people to enter crippling debt, put their body through the abuse of childbirth, and possible forced poverty because of lack of childcare or compensation for missing work isn’t okay.

Additionally, anti-abortion really only seems to be concerned with one thing - popping out children. There is ZERO concern for the health, wellbeing, or survival of that child OR the parent afterward. This is oppressive and forced childbirth expectations. And again, reduces womb-having people as nothing more than a means to an end. Their life and wellbeing aren’t considered - they’re incubators.

4) “How is this a woman’s autonomy problem.”All of the above. The entire idea of denying women normal reproductive medical procedures or criminalize a natural thing that our bodies DO is inherently oppressive. Deciding that a womb-having person is just supposed to do their best to carry to term an embryo regardless of danger to their life, medical needs, e, inability to care for the child, inability to pay medical bills, or the abuse that childbirth puts on the body… and possibly condemning them to death, poverty, or life-long debt removes the ability for a person to choose what is done or what is done TO their body. It’s inherently oppressive.

Make no mistake, these bills have very little to do with saving the lives of children, and everything to do with keeping women impoverished, oppressed, and without any control over their own bodies and lives. These bills are also written and signed without ANY input or oversight primarily by the people they affect. This is not a choice that womb-having people made… these are oppressive laws being forced upon them.

Some final personal notes from me: I am currently in a place where I would suffer greatly from these laws if they were to be implemented in my state. First of all, if I were to get pregnant, mine would be a high-risk pregnancy. It is likely that I could lose the pregnancy anytime within the first two trimesters, which would require an abortive procedure to remove the remaining tissue. If I’m to get pregnant, I need to know that modern medical procedures that are agreed to be the most effective best practices would be available to me by a doctor without the threat of criminalization or debating on whether it’s necessary/legal. This affects all people who may ever become pregnant. This is a clear and present fear for us. It’s not just anti-abortion. If that’s all it was… the answer would be simple, don’t have one. If you need one to save your life, you can choose to say no. But it’s not. This is about controlling womxn, denying us healthcare, and we are afraid. We are all desperately terrified of this becoming the new normal across our country. ONE in FOUR pregnancies ends in the need for abortion. And if you need one, you get one. This is about whether or not we have access to SAFE and MEDICALLY sanctioned abortions. 

I really encourage you to do some additional research and reading from educational sites. Be wary of both FOX News, CNN, major news networks, and any journalists with a religious agenda. Further reading: https://prochoice.org/education-and-advocacy/downloads-resources/https://iwhc.org/2018/09/abortion-normal-and-vital/https://www.prochoiceamerica.org/issue/abortion-access/I’d really appreciate if any followers could tack on additional resources, statistics, and personal stories. This is SO important. 

[myfandomrambles]

Azula Character Analysis

Born a princess with a high position

Was favored by her father and second in her mother's eyes

Deeply skilled bender from a young age

Attended the royal girl Fier Nation academy and had private tutors

Was around severe abuse emotional and occasionally physical

Abandoned by mother due to political intrigue which affected her

Aware of her parents' murder of her grandfather

Rose to crown princess after her father coronation and brother banishment which she watched

Was used as a weapon by her father against her uncle and bother by age 14

Had a few friends who became partners in hunting the avatar her brother

Was a skilled tactician able to infiltrate the heart of the earth kingdom

Overthrow the earth king

Nearly killed the avatar

Achieved high status for this but shared it with her brother is a skilled protective manner

Had a tumultuous relationship with her brother

Betrayed by her friends 

Attempted to kill her brother

Became fire lord but felt rejected through this as she was unable to claim victory along with her father

Suffered a mental health breakdown 

Was beaten by Zuko & katara in an Agni Kai losing her status 

Overview:

Azula is a complex character who is gifted, clever, beautiful, and deeply psychologically injured. Her story is one of abuse, manipulation, and war. She was raised by abusive people in a cult of power and supremacy; by the age of 14, she was being used to put the same trauma out on the entire world. 

The prime driver for Azula’s character is the necessity to retain control over her situation and due to her status as the princess of the world’s dominant power, this is control over everything. Control and power are the only things Azula truly understands as valuable. This control also equals safety, safety from physical harm during a battle, and emotional harm by others. We can see this control manifest in her emotional distress at having even one hair out of place during her training (2x01). She uses her place of power to hold fear over other people, those she considers lesser than her, by invoking the fear of losing their place and physical harm. Her social power and skill in bending back up the threats. 

Azula’s need for control started as a child who grew up being taught through the iron hand of Ozai who demanded perfection. Her status as a prodigy with fire bending, physical aptitude, and intelligence gave her positive attention from her father but also led her to be inculcated even stronger into the idea that fear is the only way. Her father taught with the fear of retribution for failure as much as any positive attention. The more blatant abuse Zuko suffered from their father for showing what was perceived as weakness and emotionality was another teacher that she must always control every part of her. (2x07, 3x06)

This control via strength without understanding was worsened by her other connections. Her mother failed to connect and attune to her daughter so even in early childhood they were always moving past each other. Azula’s failure to show empathy was met with judgment and punishment and we don’t see them ever repair the relational rupture. Their mother then abandoned them accompanied by their parents murdering their grandfather and threats against her brother. Leaving her with only Ozai as a point of influence and even more surrounded by violence. (2x07)

Azula also gained little perspective outside of the pure ideology of the fire nation royal family and royal academy for girls. She carried the beliefs of fire supremacy and nationalism with no outside input which left her with the schemas of power in her nascent socio-political awareness and added to the stunting of her ability to gain empathy. She was taught to view the world as nations and people only worth understanding to beat not for its own sake. (2x07, 2x19-20, 3x05, 3x06)

This pain leaves her spending the whole second season [spanning months] as a weapon of her father. She is forced to travel around the world, originally with only staff, with the goal of hurting her own family in the name of not being shamed. To prove she can do it she gives herself the task of stopping the avatar. Azula is able to escape a fight with some of the strongest benders we see in the show easily and is persistent to a fault. (2x03) We see her skills of strategy and combat shine here as well as many of her trauma responses. The biggest one being she is acting in a mindset that can not shift from using the world and not experiencing it. (2x01, 2x03, 2x07-8, 2x13, 2x15)

 Azula’s pure genius shows in her ability to take over Ba Sing Se on her ability to read other people, manipulate court games, and her sheer belief in her infallibility. We see her play the Dai Li and Long Feng with only the backup of her two friends. She has an iconic moment of power “Don’t flatter yourself you were never even a player” and invokes her belief in the divine right of kings or lords. (2x18-20)

Once she proves herself and can bring her brother and uncle home, if in a way not planned, putting her back in a secure place of princess she longs to keep. We see her try and maintain control by being the one who understands both her father and Zuko. We see her struggle greatly with normal life but thrive within the system of the place. (3x05-6, 3x11)

However, we see her set world start to collapse when Zuko leaves and her only two friends choose to take a chance for love versus staying in her control bubble. This challenges her sense of safety she works so hard to maintain. It also goes against her understanding of interpersonal relationships and her innate power. (3x11, 3x13, 3x15)

This causes a breakdown in the end. However, this leaves her without a throne and a sense of safety. After the show, we see her mental health stay in a deteriorated state, struggle with the past, and joins a group that wants to harm the new age of peace. (3x18-20, Comics: The Promise Part Three, The Search. Smoke and Shadow)

Relationships:

Zuko

Zuko and Azula are one of the key family dynamics within the story. Azula acts as a foil to Zuko during their childhood being the golden child to his scapegoat. She was Ozai’s favorite whereas Zuko was closer to Ursa. They both suffered severe trauma as young people but Azula spent the time trying to not be viewed as poorly as Zuko. (3x07, The Search) Something she directly tells their father, to not be treated like Zuko (3x18). Building your relationship with your sibling as wanting to prove you are better than them sets them up too but heads, something she acknowledges was also going to come down to them deciding who is the right one to succeed their father. 

During the main plot, we see them start as the predator and the prey (2x01, 2x08). Both of them lived in the mindset Ozai Taught them, she was born lucky and he was lucky to be born (1x20, 2x07). She is the long arm of their father only claiming some autonomy when she chooses her team and attacks the avatar as well (2x03, 2x08). 

Azula brings Zuko back into their fold because next to Iroh she understands Zuko the best. She knows easily the only thing he wants is to feel in control of his life and craves the respect of their father, these are things she also needs. We see her also offer the double-sided act of letting Zuko take credit. It is partially protective as should he live Azula is protected, Suko would be the one who failed. It can also be some degree of kindness for her brother because she does like the system the way it is and Zuko being in pain causes worse stress. 

They continue to bump heads as we see Azula feel most at home within the bureaucracy whereas he struggles to feel as if it was right. Zuko still carries the pain of shame for his actions at the same time Azula pushes much of her emotion down. Part of this is Azula knows where she stands and as long as others play the part she has no worries. Zuko breaks this steady normal as a child when he wants to be empathetic to soldiers and again when he feels the need to save the earth kingdom she wanted to kill in total war. (3x01, 3x05, 3x16)

Their reactions during The Day of Black Sun (3x11) set them on their paths for the end and they mirror each other. Azula uses the time to play her role and waits for the fire bending to turn on to win. Zuko uses the time to pull away from their father for good. They continue to be antagonistic and Zuko is an axle in her relational rupture with Ty Lee and Mai. (3x14-16)

Their final Agni Kai for the title of Firelord shows how much Zuko has learned in his complex bending style and ability to hold control while we notice Azula loose form entirely relying almost completely on her raw power. Her very body language giving off how sick she is currently in her movements now disjointed and lacking precision which conflicts with the controlled fighting we see from Katara and Zuko. (3x18-21).

They have spent their whole life used as pawns by their parents and stuck in the milieu of war and suffering. Azula’s status as her father’s favorite offers her the status Zuko wants but she also lacks the time and ability to grow Zuko earned through his relationship with Uncle Iroh. Their understanding of each other is strong but Azula fails to offer sympathy to her brother when he chooses things she wouldn’t and treats him poorly. And Zuko needs to be able to challenge her so he can properly heal, along with team avatar, the fissures in the world.

Ursa

We see that Azula and Ursa do not understand each other and the abuse they both suffer disallowed them to properly attach. Ursa didn’t understand Azula’s natural predispositions or her trauma which left Azula often being told off by her mother or treated as separate from the bond Ursa had with Zuko. I wouldn’t go as far as to say Ursa was forming a scapegoat golden child dynamic more so she couldn’t bring herself to look past her trauma. (2x07, The Promise, & The Search)

During the fire nation teens' conversation at ember island, we see that Azula generalized her mother's view of her as a monster as much as the other conditioning she had as a kid. This whole where her mother’s attunement should be opened even more space for what Ozai taught her. Azula lack’s a full ability to process this but it is the one time we see Azula even come close to verbalizing painful emotions other than paranoia and anger. (3x05)

If we are to believe the memory we see from Iroh (1x12) she was already immune to violence as a pre-teen believing that Ozai’s assault of Zuko was justified and even taking gratification from it. This play into her relationship with her mother as the gentleness her mother might have displayed towards her child was missing making the hardest part of the indoctrination become the most prevalent. Worsen when Ursa abandons her children and seeks out her new life. The effects of this are her willingness to be cavalier with life, and failure to attach to others (3x17).

Azula’s relationship with her mother ends up being the breaking point in space after the betrayal of her friends. When we see her experience hallucinations and paranoid thought they center around her mother and their relationship, rather there was love or not being the central question. (3x19) 

Paranoid delusions around her mother continue in the comics where we see Azula unable to interact from a clear headspace. (The Search)

Ozai

Ozai is Azula's main force of identity shaping her internal and external perceptions to the point of making her more of a human tool than a real daughter. The craving for her father's need is just as strong as Zuko’s but instead of trying to restore it her job is to keep it and not rock the boat. This is seen in her letting Zuko take credit for the killing the avatar which brings her brother back in (3x01) and when she asks Ozai to not treat her like Zuko when he becomes the phoenix king (3x20)

Throughout the show, everything she does is to please her father from going after her brother and then succeeding in killing Aang. (2x01-2x20). She also parrots her father's belief about weakness, fears power, and the might of the fire nation. Examples include naming the city New Ozai, demanding the divine right of kings, and her obsessive focus on acting and appearing perfect. 

Ozai’s abuse permeated everything Azula was and is leading to her becoming the shadow of a person we see at the end of the series. 

Ty Lee & Mai

Next to her blood family Mai & Ty Lee are her most influential relationships. She considers them generally friends starting when they went to the same school (2x07). We see that even as a child she had the highest status in the group and already needed to win. However, they do seem to have some genuine care for the princess even if it is never balanced. For example, when recruiting Ty Lee she uses manipulation and fear to force her back into serving the fire nation. (2x03). Mai and Ty Lee are skilled fighters making them useful to Azula, something she values more than anything other than loyalty. She has trouble conceptualizing their emotions as validly seen in her calling their emotions performances, however in the same episode we see her care about making Ty Lee cry and experience very human emotions of envy herself. They bond over their traumas and their shared love of destruction (3x05).

None of the three of them are particularly well adjusted but what Azula has on her side is an utter belief in her competence and her belief that their friends will fall in line with that ambition. For the most part, they do; Ty Lee often flatters her and Mail generally does as she’s told when Azula is around. However one of Azula’s most pivotal moments comes when this obedience falls through. Mai loves Zuko more than she fears Azla’s wrath and Ty Lee can’t bear to see them hurt each other. Earning one of Azula's most characterizing lines ``You should have feared me more”. This betrayal and shift in her stable world put Azula over the edge and fuels paranoid thoughts and a slip into worse mental illness. (3x14)

To consider Mai and Ty Lee to be the manipulative ones or otherwise treat them as the bad or abusive party to Azula is unfair. They are doing what they can as they believe Azula has the right to be in charge and suffer consequences when they step out of line. However, it’s equally unfair to assume everything Azula does is machiavellian; she too is acting on sincerely held beliefs and as a daughter of abusive or neglectful parents. I think Azula has a hard time conceptualizing others as full people objectifying them in her schema of the world but unlike some of her behavior to Zuko, I doubt it’s intentionally cruel. 

Developmetnal Trauma

Adulti-fication (2x01,2x03, 2x07, 2x08, 2x13 2x19-20, 3x01-2, 3x05, 3x15, & 3x18-21)

Anger (3x13-20)

Control fixations (2x01, 2x03, 2x07, 2x7, 2x13, 2x19, 3x01, 3x05, 3x15, & 3x18-21)

Conditioned Value Systems (2x01, 2x07, 2x19-20, 3x01, & 3x05)

Empathic Deficits (2x03, 2x07, 2x15, 3x05, & 3x11)

Harm to Animals (2x07 & 2x15)

Hypervigilance (2x01, 2x03, 2x07, 2x08, & 2x19-20)

Obsessive Thoughts (2x01, 2x03, 3x05, & 3x18-21)

Locus of Control breakdown (2x01,2x07,  2x15, 3x05, 3x14, & 3x18-21)

Paradoxical Arousal, [Functions best during high-stress situations and worse under normal or positive] (2x03, 3x05, 3x13-5)

Paranoid Thoughts (3x18-21)

Perfectionism (2x01, 2x07, 3x05, 3x13, & 2x17-20)

Positive & Negative Psychosis Symptoms (3x18-21)

Recklessness (2x03, 2x07, 2x08, 3x05, & 3x15-16)

Risk Seeking Behaviors (2x03, 2x13, 2x15, & 3x11)

Social issues (3x05)

Trust Issues (2x13, 3x01, 3x11, 3x13, & 3x18-21)

Violent Play Behaviors (2x07 & 3x05)

[jccamus]

Here are the 7 requirements for building ethical AI, according to the EU commission

In October, Amazon had to discontinue an artificial intelligence–powered recruiting tool after it discovered the system was biased against female applicants. In 2016, a ProPublica investigation revealed a recidivism assessment tool that used machine learning was biased against black defendants. More recently, the US Department of Housing and Urban Development sued Facebook because its ad-serving algorithms enabled advertisers to discriminate based on characteristics like gender and race. And Google refrained from renewing its AI contract with the Department of Defense after employees raised ethical concerns.

Those are just a few of the many ethical controversies surrounding artificial intelligence algorithms in the past few years. There’s a six-decade history behind the AI research. But recent advances in machine learning and neural networks have pushed artificial intelligence into sensitive domains such as hiring, criminal justice and health care.

In tandem with advances in artificial intelligence, there’s growing interest in establishing criteria and standards to weigh the robustness and trustworthiness of the AI algorithms that are helping or replacing humans in making important and critical decisions.

With the field being nascent, there’s little consensus over the definition of ethical and trustworthy AI, and the topic has become the focus of many organizations, tech companies and government institutions.

In a recently published document titled “Ethics Guidelines for Trustworthy AI,” the European Commission has laid out seven essential requirements for developing ethical and trustworthy artificial intelligence. While we still have a lot to learn as AI takes a more prominent role in our daily lives, EC’s guidelines, unpacked below, provide a nice roundup of the kind of issues the AI industry faces today.

Human agency and oversight

Image credit: Depositphotos

“AI systems should both act as enablers to a democratic, flourishing and equitable society by supporting the user’s agency and foster fundamental rights, and allow for human oversight,” the EC document states.

Human agency means that users should have a choice not to become subject to an automated decision “when this produces legal effects on users or similarly significantly affects them,” according to the guidelines.

AI systems can invisibly threaten the autonomy of humans who interact with them by influencing their behavior. One of the best-known examples in this regard is Facebook’s Cambridge Analytica scandal, in which a research firm used the social media giant’s advertising platform to send personalized content to millions of users with the aim of affecting their vote in the 2016 U.S. presidential elections.

The challenge of this requirement is that we’re already interacting with hundreds of AI systems everyday, including the content in our social media feeds, when we view trends in Twitter, when we Google a term, when we search for videos on YouTube, and more.

The companies that run these systems provide very few controls over the AI algorithms. In some cases, such as Google’s search engine, companies explicitly refrain from publishing the inner-workings of their AI algorithms to prevent manipulation and gaming. Meanwhile, various studies have shown that search results can have a dramatic influence on the behavior of users.

Human oversight means that no AI system should be able to perform its functions without some level of control by humans. This means that humans should either be directly involved in the decision-making process or have the option to review and override decisions made by an AI model.

In 2016, Facebook had to shut down the AI that ran its “Trending Topics” section because it pushed out false stories and obscene material. It then returned humans in the loop to review and validate the content the module was specifying as trending topics.

Technical robustness and safety

The EC experts state that AI systems must “reliably behave as intended while minimizing unintentional and unexpected harm, and preventing unacceptable harm” to humans and their environment.

One of the greatest concerns of current artificial intelligence technologies is the threat of adversarial examples. Adversarial examples manipulate the behavior of AI systems by making small changes to their input data that are mostly invisible to humans. This happens mainly because AI algorithms work in ways that are fundamentally different from the human brain.

Adversarial examples can happen by accident, such as an AI system that mistakes sand dunes for nudes. But they can also be weaponized into harmful adversarial attacks against critical AI systems. For instance, a malicious actor can change the coloring and appearance of a stop sign in a way that will go unnoticed to a human but will cause a self-driving car to ignore it and cause a safety threat.

Adversarial attacks are especially a concern with deep learning, a popular blend of AI that develops its behavior by examining thousands and millions of examples.

There are already been several efforts to build robust AI systems that are resilient to adversarial attacks. AutoZOOM, a method developed by researchers at MIT-IBM Watson AI Lab, helps detect adversarial vulnerabilities in AI systems.

The EC document also recommends that AI systems should be able to fallback from machine learning to rule-based systems or ask for a human to intervene.

Since machine learning models are based on statistics, it should be clear how accurate a systems is. “When occasional inaccurate predictions cannot be avoided, it is important that the system can indicate how likely these errors are,” the EC’s ethical guidelines state. This means that the end user should know about the confidence level and the general reliability of the AI system they’re using.

Privacy and data governance

“AI systems must guarantee privacy and data protection throughout a system’s entire lifecycle. This includes the information initially provided by the user, as well as the information generated about the user over the course of their interaction with the system,” according to the EC document.

Machine learning systems are data-hungry. The more quality data they have, the more accurate they become. That’s why companies have a tendency to collect more and more data from their users. Companies like Facebook and Google have built economic empires by building and monetizing comprehensive digital profiles of their users. The use this data to train their AI models to provide personalized content and ads to their users and keep them glued to their apps to maximize their profit.

But how responsible are these companies in maintaining the security and privacy of this data? Not very much. They’re also not very explicit about the amount of data they collect and ways they use it.

In recent years, general awareness about privacy and new rules such as the European Union’s General Data Protection Regulation (GDPR) and California’s Consumer Privacy Act (CCPA) are forcing organizations to be more transparent about their data collection and processing practices. In the past year, many companies have offered users the option to download their data or to ask the company to delete it from its servers.

However, more needs to be done. Many companies share sensitive user information with their employees or third-party contractors to label data and train their AI algorithms. In many cases, users don’t know that human operators review their information and they falsely believe that only algorithms process their data.

Very recently, Bloomberg revealed that thousands of Amazon employees across the world access the voice recordings of the users of its Echo smart speakers to help improve the company’s AI-powered digital assistant Alexa. The idea does not sit well with the users, who expect to enjoy privacy in their homes.

Transparency

The European Commission experts define AI transparency in three components: traceability, explainability and communication.

AI systems based on machine learning and deep learning are highly complex. They develop their behavior based on correlations and patterns found in thousands and millions of training examples. Often, the creators of these algorithms don’t know the logical steps behind the decisions their AI models make. This makes it very hard to find the reasons behind the errors these algorithms make.

EC specifically recommends that developers of AI systems document the development process, the data they use to train their algorithms, and explain their automated decisions in ways that are understandable to humans.

Explainable AI has become the focus of several initiatives by the private and public sector. This includes a widespread effort by the Defense Advanced Research Projects Agency (DARPA) to create AI models are open to investigation and methods that can explain AI decisions.

Another important point raised in the EC document is communication. “AI systems should not represent themselves as humans to users; humans have the right to be informed that they are interacting with an AI system,” the document reads.

Last year, Google introduced Duplex, an AI service that could place calls on behalf users and make restaurant and salon reservations. Controversy ensued because the assistant refrained from presenting itself as an AI agent and duped its interlocutors into thinking they were speaking to a real human. The company later updated the service to present itself as Google Assistant.

Diversity, non-discrimination and fairness

Algorithmic bias is one of the well-known controversies of contemporary AI technology. For a long time, we believed that AI would not make subjective decisions based on bias. But machine learning algorithms develop their behavior from their training data, and they reflect and amplify any bias contained in those data sets.

There have been numerous examples of algorithmic bias rearing its ugly head, such as the examples listed at the beginning of this article. Other cases include a study that showed popular AI-based facial analysis servicesbeing more accurate on men with light skin and making more errors on women with dark skin.

To prevent unfair bias against certain groups, EC’s guidelines recommend that AI developers make sure their AI systems’ data sets are inclusive.

The problem is, AI models often train on data that is publicly available, and this data often contains hidden biases that already exist in the society.

For instance, a group of researchers at Boston University discovered that word embedding algorithms (AI models used in tasks such as machine translation and online text search) trained on online articles had developed hidden biases, such as associating programming with men and homemaker with women. Likewise, if a company trains its AI-based hiring tools with the profiles of its current employees, it might be unintentionally pushing its AI toward replicating the hidden biases and preferences of its current recruiters.

To solve hidden biases, EC recommends for companies that develop AI systems hire people from diverse backgrounds, cultures and disciplines.

One consideration to note however is that fairness and discrimination often depends on the domain. For instance, in hiring, organizations must make sure that their AI systems don’t make decisions. But in another field like health care, parameters like gender and ethnicity must be factored in when diagnosing patients.

Societal and environmental well-being

“[The] broader society, other sentient beings and the environment should be also considered as stakeholders throughout the AI system’s life cycle,” EC’s guidelines state.

The social aspect of AI has been deeply studied. A notable example are social media companies, which use AI to study the behavior of their users and provide them with personalized content. This makes social media applications addictive and profitable, but also causes a negative impact on users, making them less social, less happy and less tolerant toward opposing views and opinions.

Some companies have started to acknowledge this and correct the situation. In 2018, Facebook declared that it would be making changes to its News Feed algorithm and provide users with more posts from friends and family and less from brands and publishers. The move was aimed at making the experience more social.

The environmental impact of AI is less discussed, but is equally important. Training and running AI systems in the cloud consumes a lot of electricity and leaves a huge carbon footprint. This is a problem that will grow worse as more and more companies use AI algorithms in their applications.

One of the solutions is to use lightweight edge AI solutions that require very little power and  run on renewable energy. Another solution is to use AI itself to help improve the environment. For instance, machine learning algorithms can help manage traffic and public transport to reduce congestion and carbon emissions.

Accountability

Finally, EC calls for mechanisms “to ensure responsibility and accountability for AI systems and their outcomes, both before and after their development, deployment and use.” Basically, this means there should be legal safeguards to make sure companies keep their AI systems conformant with ethical principles.

U.S. lawmakers recently introduced the Algorithmic Accountability Act which, if passed, will required companies to have their AI algorithms evaluated by the Federal Trade Commission for known problems such as algorithmic bias as well as privacy and security concerns.

Other countries, including the UK, France and Australia have passed similar legislation to hold tech companies to account for the behavior of their AI models.

In most cases, ethical guidelines are not in line with the business model and interests of tech companies. That’s why there should be oversight and accountability. “When unjust adverse impact occurs, accessible mechanisms should be foreseen that ensure adequate redress. Knowing that redress is possible when things go wrong is key to ensure trust,” the EC document states.

This story is republished from TechTalks, the blog that explores how technology is solving problems… and creating new ones. Like them on Facebook here and follow them on Twitter.

TNW Conference 2019 is coming! Check out our glorious new location, inspiring line-up of speakers and activities, and how to be a part of this annual tech extravaganza by clicking here.

Publicado en The Next Web http://bit.ly/2J8JDzA vía IFTTT