#azure chatgpt 4

[tsrtimes]

Simplifying Testing Infrastructure with Cloud Automation Testing

In today’s fast-paced digital world, businesses need to continuously deliver high-quality software applications to meet customer expectations. But how are businesses going to make sure whether their product meets the highest functionality, usability, security and performance standards? This is where Software Testing comes into the picture and ensures the performance and quality of the product.

There are two methods of testing: manual and automated. However, manual testing is time-consuming and can be prone to errors. With the rise in the scope and scale of testing in DevOps, the requirement for automated testing has become apparent.

What is Automation Testing? 

Automation Testing is the process of automatically running tests. They are used to implement text scripts on a software application. These tests are conducted with the help of testing software which frees up resources and time during the process of testing. This enables people to test the quality of the software more judiciously and at a lower cost.

Automation testing allows people to:

Create a set of tests that can be reused multiple times

Save cost by debugging and detecting problems earlier

Deploy tests continuously to accelerate product launches

How Automation Testing is Transforming the World? 

Automation can be seen almost everywhere, not only in QA testing but in our day-to-day lives. From self-driving cars to voice tech, the technology is rapidly becoming automated to simplify our lives.

Automation testing is consistently improving the quality of QA testing and saves a lot of time in comparison to manual testing. As known, writing test cases need continuous human intervention. And to ensure the best results, test cases should be a continuous collaboration between testers and developers.

No matter the product or service, the key benefits of automation testing can be summarized as the following:

Increased speed

Increased output

Enhanced quality

Lesser cost

Advantages of Automation Testing 

With the improvement in AI, the power and scope of automated testing tools are increasing rapidly. Let’s look into detail as to what people and organizations can gain from automation testing:

Saves cost 

Automated Software Testing will help your business save time, money and resources during the process of quality assurance. While there will be a requirement for manual testing too, the QA engineers will have the time to invest in other projects. This will lower the overall cost of software development.

Simultaneously run test 

Since automated testing needs little to no human intervention once it starts, it becomes easy to run multiple tests at once. This also provides you with the opportunity to create comprehensive comparative reports faster with the same parameters.

Quicker feedback cycle 

In the case of manual tests, it can take a lot of time for testers to return to your DevOps department with feedback. Using automation testing tools, you can implement quicker validation during the software development process. By testing at the earlier stages, you increase the efficiency of your team.

Lesser time to market 

The time that is saved with continuous testing during development contributes to the earlier launch of your product. Automation testing tools can also enable faster test results, speeding up final software validation.

Improved test coverage  

With a well-planned automation strategy, you can expand your test coverage to improve the quality of greater features in your application. Due to the automation of the testing process, it gives free time to your automation engineers to write more tests and make them more detailed.

Better insights 

Automation tests not just disclose when your test fails but also reveal application insights. It can demonstrate data tables, file contents, and memory contents. This will allow developers to identify what went wrong.

Enhanced accuracy 

Making mistakes is human and in the case of manual testing, there’s always a possibility of human error. Whereas with automation, the implementation of tests will be accurate most of the time. Obviously, test scripting is done by humans, which implies that there’s still some risk of making errors. But these errors will become lesser and lesser the more you reuse tests.

Lesser stress on your QA team 

Your quality assurance team will experience significantly less stress if you adopt an automated testing technique. Once you eliminate the hassle of manual testing, you give them the time to create tools that improve your testing suite even further.

Types of Automated Testing 

Unit Testing If the individual parts of your code won’t function correctly, there is no possibility for them to work within the final product. Unit testing looks into the smallest bit of code that can be segregated in a system. To conduct Unit Tests, the tester must be aware of the internal structure of the program. The best thing about Unit Testing is that it can be implemented throughout the process of software development. This ensures consistent delivery of feedback that will speed up the development process, sending products to market faster.

Functional Testing

After ensuring that all the individual parts work, you need to check whether the system functions based on your specifications and requirements. Functional Testing makes sure that your application works as it was planned based on requirements and specifications. Functional Testing assesses the APIs, user interface, security, database and other functionalities.

Regression Testing 

Regression tests are required to confirm that a recent change made in the system hasn’t impacted the existing features. To perform these tests, you extract current relevant test cases from the test suite that involves the affected and modified parts of the code. You must carry out regression testing whenever you modify, alter or update any part of the code.

Load Testing 

Do you know how much pressure your application can take? This is an essential piece of information to keep yourself prepared with before you hand it over to your user. Load tests are non-functional software tests which are carried out to test your software under a specified load. It will demonstrate the behaviour of the software while being put under the stress of various users.

Performance Testing 

Performance Testing assesses the responsiveness, stability, and speed of your application. If you don’t put your product through some sort of performance test, you’ll never know how it will function in a variety of situations.

Integration Testing

Integration testing involves testing how the individual units or components of the software application work together as a whole. It is done after unit testing to ensure that the units integrate and function correctly.

Security Testing

Security testing is used to identify vulnerabilities and weaknesses in the software application’s security. It involves testing the application against different security threats to ensure that it is secure.

GUI Testing

GUI testing involves testing the graphical user interface of the software application to ensure that it is user-friendly and functions as expected.

API Testing

API testing involves testing the application programming interface (API) to ensure that it functions correctly and meets the requirements.

Choosing a Test Automation Software Provider 

If your business is planning to make the move, the test automation provider you pick should be able to provide:

Effortless integration with CI/CD pipeline to facilitate automation, short feedback cycle and fast delivery of software.

The capability to function on private or public cloud networks.

Integration with the current infrastructure for on-site testing for simpler test handling and reporting.

Remote access to real-time insights and monitoring tools that can help you better understand user journeys and how a certain application is being used.

Automated exploratory testing to increase application coverage.

Test environments that are already set up and can be quickly launched when needed.

CloudScaler: A Trusted Provider of Software Testing in the Netherlands

With the increasing complexity of modern software development, the need for reliable and efficient software testing services has never been greater. CloudScaler, a trusted provider of Software Testing in the Netherlands, offers a comprehensive suite of testing services to help teams navigate the challenges of cloud infrastructure and microservices.

Our services are designed to shorten deployment times and development costs, enabling your team to focus on what they do best: building innovative software solutions. Our approach is rooted in efficiency, reliability, and expertise, ensuring that you can trust CloudScaler as your partner in software testing.

[jcmarchi]

ChatGPT-4 vs. Llama 3: A Head-to-Head Comparison

New Post has been published on https://thedigitalinsider.com/chatgpt-4-vs-llama-3-a-head-to-head-comparison/

ChatGPT-4 vs. Llama 3: A Head-to-Head Comparison

As the adoption of artificial intelligence (AI) accelerates, large language models (LLMs) serve a significant need across different domains. LLMs excel in advanced natural language processing (NLP) tasks, automated content generation, intelligent search, information retrieval, language translation, and personalized customer interactions.

The two latest examples are Open AI’s ChatGPT-4 and Meta’s latest Llama 3. Both of these models perform exceptionally well on various NLP benchmarks.

A comparison between ChatGPT-4 and Meta Llama 3 reveals their unique strengths and weaknesses, leading to informed decision-making about their applications.

Understanding ChatGPT-4 and Llama 3

LLMs have advanced the field of AI by enabling machines to understand and generate human-like text. These AI models learn from huge datasets using deep learning techniques. For example, ChatGPT-4 can produce clear and contextual text, making it suitable for diverse applications.

Its capabilities extend beyond text generation as it can analyze complex data, answer questions, and even assist with coding tasks. This broad skill set makes it a valuable tool in fields like education, research, and customer support.

Meta AI’s Llama 3 is another leading LLM built to generate human-like text and understand complex linguistic patterns. It excels in handling multilingual tasks with impressive accuracy. Moreover, it’s efficient as it requires less computational power than some competitors.

Companies seeking cost-effective solutions can consider Llama 3 for diverse applications involving limited resources or multiple languages.

Overview of ChatGPT-4

The ChatGPT-4 leverages a transformer-based architecture that can handle large-scale language tasks. The architecture allows it to process and understand complex relationships within the data.

As a result of being trained on massive text and code data, GPT-4 reportedly performs well on various AI benchmarks, including text evaluation, audio speech recognition (ASR), audio translation, and vision understanding tasks.

Text Evaluation

Vision Understanding

Overview of Meta AI Llama 3:

Meta AI’s Llama 3 is a powerful LLM built on an optimized transformer architecture designed for efficiency and scalability. It is pretrained on a massive dataset of over 15 trillion tokens, which is seven times larger than its predecessor, Llama 2, and includes a significant amount of code.

Furthermore, Llama 3 demonstrates exceptional capabilities in contextual understanding, information summarization, and idea generation. Meta claims that its advanced architecture efficiently manages extensive computations and large volumes of data.

Instruct Model Performance

Instruct Human evaluation

Pre-trained model performance

ChatGPT-4 vs. Llama 3

Let’s compare ChatGPT-4 and Llama to better understand their advantages and limitations. The following tabular comparison underscores the performance and applications of these two models:

Aspect ChatGPT-4 Llama 3 Cost Free and paid options available Free (open-source) Features & Updates Advanced NLU/NLG. Vision input. Persistent threads. Function calling. Tool integration. Regular OpenAI updates. Excels in nuanced language tasks. Open updates. Integration & Customization API integration. Limited customization. Suits standard solutions. Open-source. Highly customizable. Ideal for specialized uses. Support & Maintenance Provided by OpenAl through formal channels, including documentation, FAQs, and direct support for paid plans. Community-driven support through GitHub and other open forums; less formal support structure. Technical Complexity Low to moderate depending on whether it is used via the ChatGPT interface or via the Microsoft Azure Cloud. Moderate to high complexity depends on whether a cloud platform is used or you self-host the model. Transparency & Ethics Model card and ethical guidelines provided. Black box model, subject to unannounced changes. Open-source. Transparent training. Community license. Self-hosting allows version control. Security OpenAI/Microsoft managed security. Limited privacy via OpenAI. More control via Azure. Regional availability varies. Cloud-managed if on Azure/AWS. Self-hosting requires its own security. Application Used for customized AI Tasks Ideal for complex tasks and high-quality content creation

Ethical Considerations

Transparency in AI development is important for building trust and accountability. Both ChatGPT4 and Llama 3 must address potential biases in their training data to ensure fair outcomes across diverse user groups.

Additionally, data privacy is a key concern that calls for stringent privacy regulations. To address these ethical concerns, developers and organizations should prioritize AI explainability techniques. These techniques include clearly documenting model training processes and implementing interpretability tools.

Furthermore, establishing robust ethical guidelines and conducting regular audits can help mitigate biases and ensure responsible AI development and deployment.

Future Developments

Undoubtedly, LLMs will advance in their architectural design and training methodologies. They will also expand dramatically across different industries, such as health, finance, and education. As a result, these models will evolve to offer increasingly accurate and personalized solutions.

Furthermore, the trend towards open-source models is expected to accelerate, leading to democratized AI access and innovation. As LLMs evolve, they will likely become more context-aware, multimodal, and energy-efficient.

To keep up with the latest insights and updates on LLM developments, visit unite.ai.

[abhishektorgalblogs]

Open Ai case study (will Ai replace humans)

*Form last few months people are talking about chat gpt but no one talking about "OpenAi" the company which made chat gpt*

After Open ai launch chat gpt, chat gpt crossed the mark of 1 million users in 5 days and 100 million at 2 month its very hard to cross thise numbers for an any other  company but Open Ai is not a like any other company it is one of game changer companies

How? You will get soon

One of the big company in the tech industry Microsoft itself invested $1billion in 2019 and On January 23, 2023, Microsoft announced a new multi-year, multi-billion doller investment in open Ai, reported to be $10 billion  . The investment is believed to be a part of Microsoft's efforts to integrate OpenAI's ChatGPT into the Bing search engine.

After thise announcement  launch of Chat gpt

even threaten the shark of the industry Google, Google is ruling this industry since 2 and half decades

But after the announcement of Microsoft

the monopoly of Google has been threaten by the direct competitor of the company

The CEO of Google sunder pichai even announced "code red" Whenever there is a major issue at Google, Google announcess"Code Red." In the company When you hear that at Google, it means "all hands on deck," thise is time to work hard

But thise is still the tip of the ice burg

Open Ai is making more project like chat gpt or better than that

*Point which I am going to cover in this post*

1.what is open ai?

2.active projects of open ai.

3.how open ai making revenue?

1.What is open ai ?

OpenAI is an American based AI research laboratory .OpenAI conducts AI research with the declared intention of promoting and developing a friendly AI. OpenAI systems run on an Azure-based supercomputing which provided by Microsoft.OpenAI was founded in 2015 by a group of high-profile figures in the tech industry, including Elon Musk,Sam Altman, Greg Brockman, Ilya Sutskever, John Schulman, and Wojciech Zaremba. Ya you heard right Elon Musk the man who says against Ai in his interviews and already invested $100million at open Ai in its initial days 

It's like saying smoking is injurious to health while smoking weed

However, Musk resigned from the board in 2018 but remained a donor

Musk in another tweet clarified that he has no control or ownership over OpenAI .

It's not because the "ai is dangerous" thing it becouse conflict between co founder and other reasons

The current CEO of Open ai is sam altman and current CTO of the company is Mira murati ya a one more Indian origin person who is running big tech company

The organization was founded with the goal of developing artificial intelligence technology that is safe and beneficial for humanity, and the founders committed $1 billion in funding to support the organization's research.

First of all you should understand that Open ai is not a company like others

It's doesn't start like others

No one made this company in their basement or something the story of open in not inspiring like others but it's valuable it's an example of

what's happens when the top tech giant's and top tech scientist create something together

2. Active projects of open ai

GPT-4: OpenAI's most recent language model, which is capable of generating human-like language and has been used for a wide range of applications, including chatbots, writing assistance, and even creative writing.

​DALL-E: a neural network capable of generating original images from textual descriptions, which has been used to create surreal and whimsical images.

CLIP: a neural network that can understand images and text together, and has been used for tasks such as image recognition, text classification, and zero-shot learning.

Robotics: OpenAI is also working on developing advanced robotics systems that can perform complex tasks in the physical world, such as manipulating objects and navigating environments.

Multi-agent AI: OpenAI is also exploring the field of multi-agent AI, which involves developing intelligent agents that can work together to achieve common goals. This has applications in fields such as game theory, economics, and social science

Developers can use the API of openAi to create apps for customer service, chatbots, and productivity, as well as tools for content creation, document searching, and more, with many providing great utility for businesses

We can Develop and deploy intelligent chatbots that can interact with customers, answer questions, and provide personalized recommendations based on user preferences.

3.how open ai making revenue ?

OpenAI is a research organization that develops artificial intelligence in a variety of fields, such as natural language processing, robotics, and deep learning. The organization is primarily funded by a combination of private investors, including Microsoft and Amazon Web Services, and research partnerships with various organizations.

OpenAI generates revenue through several means,

including: AI products and services: OpenAI offers a range of AI products and services to businesses and organizations, including language models, machine learning tools, and robotic systems.

Research partnerships: OpenAI collaborates with businesses, governments, and academic institutions on research projects and consultancies.

Licensing agreements: OpenAI licenses its technologies and patents to third-party companies and organizations, allowing them to use OpenAI's technology in their own products and services.

Investments: OpenAI has received significant investments from various companies and organizations, which have provided the organization with funds to support its research and development efforts

When the open ai started by Elon Musk and other founder in 2015 the open ai started as non profit organization

but right now open ai is not properly non profit organization

It has it's for-profit subsidiary corporation OpenAI Limited Partnership (OpenAI LP)

and it's non profit organization called OpenAI Incorporated (OpenAI Inc.)

.

If it's continuous like thise open ai will play good role in technology

but the main question is

Will Ai replace humans?

It is unlikely that AI will completely replace humans in the foreseeable future. While AI has made significant advances in recent years, there are still many areas where humans excel and where machines struggle to match human performance.

AI is particularly good at performing repetitive tasks and processing large amounts of data quickly, but it lacks the creativity, empathy, and emotional intelligence that humans possess. Additionally, AI is only as good as the data it is trained on, and biases in the data can lead to biased AI systems.

Furthermore, many jobs require human-to-human interaction, which AI cannot replicate. For example, jobs in healthcare, education, and social work require empathy, understanding, and interpersonal skills, which machines are not capable of.

Overall, it is more likely that AI will augment human abilities rather than replace them entirely. As AI technology continues to develop, we may see more and more tasks being automated, but there will always be a need for human oversight and decision-making.

But there is a chance that Ai will definitely replace you

If you did not upgrade your self

If you still in bottom of your company ai will definitely replace you and I am not just talking about company I am talking about overall in general aspects it might be artist,coder,writer, editor,content creator, labour, farmer's ect.

If you are not upgrade your self you will gona replaced by Ai and machine,s

So upgrade your self using ai and make your place in the upcoming world of Ai and machine

"Ai will never replace humans but the humans who use ai they replace humans who don't use Ai"

There is evil side of Ai also but if we can create Ai than we can also create things to deal with ai

It could be anything it could be regulation's or any terms and conditions

but the point is we can use Ai to do are thing

easy

[tastydregs]

GPT-4 will hunt for trends in medical records thanks to Microsoft and Epic

Enlarge / An AI-generated image of a pixel art hospital with empty windows.

Benj Edwards / Midjourney

On Monday, Microsoft and Epic Systems announced that they are bringing OpenAI's GPT-4 AI language model into health care for use in drafting message responses from health care workers to patients and for use in analyzing medical records while looking for trends.

Epic Systems is one of America's largest health care software companies. Its electronic health records (EHR) software (such as MyChart) is reportedly used in over 29 percent of acute hospitals in the United States, and over 305 million patients have an electronic record in Epic worldwide. Tangentially, Epic's history of using predictive algorithms in health care has attracted some criticism in the past.

In Monday's announcement, Microsoft mentions two specific ways Epic will use its Azure OpenAI Service, which provides API access to OpenAI's large language models (LLMs), such as GPT-3 and GPT-4. In layperson's terms, it means that companies can hire Microsoft to provide generative AI services for them using Microsoft's Azure cloud platform.

The first use of GPT-4 comes in the form of allowing doctors and health care workers to automatically draft message responses to patients. The press release quotes Chero Goswami, chief information officer at UW Health in Wisconsin, as saying, "Integrating generative AI into some of our daily workflows will increase productivity for many of our providers, allowing them to focus on the clinical duties that truly require their attention."

The second use will bring natural language queries and "data analysis" to SlicerDicer, which is Epic's data-exploration tool that allows searches across large numbers of patients to identify trends that could be useful for making new discoveries or for financial reasons. According to Microsoft, that will help "clinical leaders explore data in a conversational and intuitive way." Imagine talking to a chatbot similar to ChatGPT and asking it questions about trends in patient medical records, and you might get the picture.

GPT-4 is a large language model (LLM) created by OpenAI that has been trained on millions of books, documents, and websites. It can perform compositional and translation tasks in text, and its release, along with ChatGPT, has inspired a rush to integrate LLMs into every type of business, whether appropriate or not.

[techinfographic]

Step-by-Step Guide to Building a Generative AI Model from Scratch

Generative AI is a cutting-edge technology that creates content such as text, images, or even music. Building a generative AI model may seem challenging, but with the right steps, anyone can understand the process. Let’s explore steps to build a generative AI model from scratch.

1. Understand Generative AI Basics

Before starting, understand what generative AI does. Unlike traditional AI models that predict or classify, generative AI creates new data based on patterns it has learned. Popular examples include ChatGPT and DALL·E.

2. Define Your Goal

Identify what you want your model to generate. Is it text, images, or something else? Clearly defining the goal helps in choosing the right algorithms and tools.

Example goals:

Writing stories or articles

Generating realistic images

Creating music

3. Choose the Right Framework and Tools

To build your AI model, you need tools and frameworks. Some popular ones are:

TensorFlow: Great for complex AI models.

PyTorch: Preferred for research and flexibility.

Hugging Face: Ideal for natural language processing (NLP).

Additionally, you'll need programming knowledge, preferably in Python.

4. Collect and Prepare Data

Data is the backbone of generative AI. Your model learns patterns from this data.

Collect Data: Gather datasets relevant to your goal. For instance, use text datasets for NLP models or image datasets for generating pictures.

Clean the Data: Remove errors, duplicates, and irrelevant information.

Label Data (if needed): Ensure the data has proper labels for supervised learning tasks.

You can find free datasets on platforms like Kaggle or Google Dataset Search.

5. Select a Model Architecture

The type of generative AI model you use depends on your goal:

GANs (Generative Adversarial Networks): Good for generating realistic images.

VAEs (Variational Autoencoders): Great for creating slightly compressed data representations.

Transformers: Used for NLP tasks like text generation (e.g., GPT models).

6. Train the Model

Training involves feeding your data into the model and letting it learn patterns.

Split your data into training, validation, and testing sets.

Use GPUs or cloud services for faster training. Popular options include Google Colab, AWS, or Azure.

Monitor the training process to avoid overfitting (when the model learns too much from training data and fails with new data).

7. Evaluate the Model

Once the model is trained, test it on new data. Check for:

Accuracy: How close the outputs are to the desired results.

Creativity: For generative tasks, ensure outputs are unique and relevant.

Error Analysis: Identify areas where the model struggles.

8. Fine-Tune the Model

Improvement comes through iteration. Adjust parameters, add more data, or refine the model's architecture to enhance performance. Fine-tuning is essential for better outputs.

9. Deploy the Model

Once satisfied with the model’s performance, deploy it to real-world applications. Tools like Docker or cloud platforms such as AWS and Azure make deployment easier.

10. Maintain and Update the Model

After deployment, monitor the model’s performance. Over time, update it with new data to keep it relevant and efficient.

Conclusion

Building a generative AI model from scratch is an exciting journey that combines creativity and technology. By following this step-by-step guide, you can create a powerful model tailored to your needs, whether it's for generating text, images, or other types of content.

If you're looking to bring your generative AI idea to life, partnering with a custom AI software development company can make the process seamless and efficient. Our team of experts specializes in crafting tailored AI solutions to help you achieve your business goals. Contact us today to get started!

[moko1590m]

セキュリティと利便性を両立させた「生成AI活用環境」の実現方法──日立ソリューションズとAllganize Japanが示す実践的アプローチ 掲載日 2024/11/18 10:00

著者：周藤瞳美

企業の生成AI活用が本格化するなか、秘匿性の高い機密情報や技術情報を安全に扱う方法に注目が集まっている。2024年10月18日に開催された「生成AIで秘匿性の高い機密情報や技術情報を安全に扱う方法とは？」と題したオンラインセミナーでは、日立ソリューションズとAllganize Japanの2社による講演が行われた。SaaS環境やプライベート環境でのセキュアな生成AI活用事例が紹介され、セキュリティと利便性を両立させた生成AI活用の具体的な道筋が示された。

生成AIは、試行から本格活用のフェーズへ 最初のセッションでは、株式会社日立ソリューションズ クラウドソリューション本部 企画部 兼AIトランスフォーメーション推進本部 AX戦略部 担当部長（現：シニアAIビジネスストラテジスト） 北林 拓���氏によって、生成AI市場の最新動向と同社の取り組みが紹介された。

国内企業における生成AI活用は、2023年度の試行フェーズを経て加速。そして2024年は本格活用のフェーズに入ったといえる。 生成AI活用の方向性として、北林氏は「攻め」と「守り」の二面性を挙げる。攻めの側面では業務効率化やサービスの高度化をめざし、守りの側面では著作権やプライバシー侵害、情報漏洩などのリスク対策が重要だという。

これらの取り組みは段階的に進められ、組織の一部での試行から始まり、全社活用とユースケース創出を経て、業務プロセス変革、そしてサービスの高度化へと発展していく。北林氏は「各フェーズの取り組みを推進するにあたってさまざまな課題があり、その課題への対処が必要になります」と話す。

続けて、北林氏はグローバルトレンドについて、北米最大級のAIカンファレンス「Ai4 2024」での知見を共有した。特に注目すべき点として、ユースケースに応じた適切なモデルを選択するマルチモデル対応の重要性が挙げられる。また、コスト効率を考慮した特定分野向けの小規模言語モデル（SLM）の活用が進んでいるという。さらに、責任あるAIの実現とリスク管理の観点から、AIのガバナンスに対する重要性が高まっていることを明らかにした。

日立ソリューションズは、2024年4月にAIトランスフォーメーション推進本部を設立。同社は、生成AIをはじめとしたAI技術を駆使するAIトランスフォーメーション（AX）を進めることにより、社会と顧客と自社のDXを加速させ、持続可能な社会の実現に向けたサステナビリティトランスフォーメーション（SX）に貢献することをめざしている。同本部の具体的な取り組みとして、ソリューションの高度化、社内業務の効率化、開発業務の効率化、そしてリスク管理・ガバナンスの4つの柱を掲げている。 実践的な活用事例も着実に成果を上げている。プロモーション業務では、従来1カ月以上かかっていたコラム作成の期間を1日程度にまで短縮することに成功した。その他、問い合わせ対応業務の効率化や、会議議事録作成の自動化、協創活動におけるアイデア発想支援など、幅広い領域で活用が進んでいるという。「今後は自社商材への生成AI適用プロジェクトも進めていきます」と北林氏は語り、継続的な取り組みの展開を示した。

同社の実践例は、生成AIの企業活用における具体的なロードマップを示すものとして注目すべき取り組みといえるだろう。

機密情報もフル活用。生成AIを「すぐに」「セキュアに」企業活用する術とは？ 続いてのセッションでは、Allganize Japan株式会社 Solution Sales Senior Manager 池上 由樹氏が登壇。生成AIの実用的な企業活用についての解説が行われた。

生成AIの企業活用において、池上氏は2つの主要な課題を指摘する。

「1つ目は、ChatGPTをはじめ生成AIを利用できる環境を全社に展開しても、従業員からすると具体的な使い方がわからないという活用面における課題、そして2つ目は、セキュリティ面での懸念です」（池上氏）

これらの課題に対し、同社はオールインワン生成AI・LLMアプリプラットフォーム「Alli LLM App Market」を提供している。 同サービスについて、池上氏は「プロンプトを入力しなくても、選ぶだけで使える生成AI・LLMアプリケーションを100個以上用意している」と説明。さらに、ノーコードでのアプリケーション作成・カスタマイズ機能、自社データとの連携機能などを備えており、企業のニーズに応じた柔軟な活用が可能となっている。

特に注目すべき機能として、企業向けに特化したRAG（Retrieval-Augmented Generation）技術を使用した機能がある。 この機能では、質問に対する回答を社内文書の中から自動で生成する際に、同社独自のRAG技術により、質問に関連する文書内の該当箇所を自動でハイライト表示し、生成された回答の根拠を明確に示すことができる。「表や画像を含む複雑な文書でも、自動で適切な前処理を行い、高精度な回答生成を実現します」と池上氏は述べる。

Alli LLM App Marketは、企業のセキュリティポリシーに応じて下記の3つの提供形態を用意している。

1.SaaS型：クラウドベースで迅速な導入が可能 最も導入が容易な形態として、SaaS型のサービスを提供している。池上氏によると、「低コストで最短1日で利用可能」という即効性が最大の特徴だ。ドキュメントなどのデータは同社が管理する環境にアップロードされ、LLMは同社が契約しているLLMのAPIサービスを経由して利用する形となる。LLMをAPIで利用する場合は、顧客企業のデータがモデル学習に利用されることはない。また、顧客企業が独自に契約しているAzure OpenAI ServiceなどのLLMと接続することも可能だ。セキュリティポリシーによってはアップロード可能なファイルに制限が生じるという課題はあるものの、迅速な導入を重視する企業には最適な選択肢となる。

2.プライベートクラウド型：セキュアな環境での運用 「最近では、プライベートクラウドを使用するケースが増えています」と池上氏は言及する。この形態では、Alli LLM App Marketをプライベートクラウド上に実装し、顧客企業が契約しているLLMと接続する構成を採用している。完全なローカル環境ではないものの、プライベートクラウド上でのデータ管理が許容される企業にとって、バランスの取れた選択肢となる。

3.オンプレミス型：完全なローカル環境での展開 最も厳格なセキュリティを必要とする企業向けの選択肢として、完全なオンプレミス環境での導入も可能だ。「特に金融機関や官公庁、製���業などからの需要が高いです」と池上氏は説明する。この形態では、LLMを含むすべてのコンポーネントを顧客企業の環境内に実装する。GPTのような大規模モデルはサイズの問題でオンプレミスには置けないため、同社が提供する専用のオンプレミスLLMを使用する形となる。なお、顧客企業が契約する特定のLLMとの接続も可能だ。

具体例として、大手証券会社での導入事例が紹介された。およそ300種類の業務マニュアルに対する高度な検索機能の実現と、生成AI活用による業務自動化基盤の構築を約3カ月で実現したという。また、クラウドサービスの利用が制限される大手銀行での完全オンプレミス環境での導入例も示された。

「企業によって求められるセキュリティ水準はさまざまです。Alli LLM App Marketはそれぞれの要件に応じた提供形態で、生成AIの活用を短期間で実現可能です」（池上氏）

導入事例から見えてきた！全社で使える安全な生成AI環境の実現方法 最後のセッションでは、2017年からAIによる業務効率化の支援に携わってきたという株式会社日立ソリューションズ スマートワークソリューション本部 ビジネス創生部 エバンジェリスト 小林 大輔氏が登壇。Alli LLM App Marketを100社以上に提案した経験をもとに、安全かつ効果的な全社展開のポイントについて解説された。

まず小林氏は、企業の生成AI活用の実態について言及。帝国データバンク「生成AIの活用に関する日本企業の最新トレンド分析（2024年9月）」によると、生成AIを「活用している」企業は全体の17.3%にとどまり、約半数の企業が「活用しておらず活用予定もない」状況だ。しかし、小林氏は「実際に活用している企業の9割近くが効果を実感しています」という。

活用部門としては経営企画部門が最も多く、企業の中枢での利用が進んでいる。また、企業規模別では1,000人以上の大企業での活用が進む一方で、小規模な企業の方が効果をより実感している傾向が見られる。それは、現状では特定の個人による利用が中心となっており、全社的な展開には至っていない企業が多いためだという。

企業内での生成AI活用を広げるための課題として、小林氏は法規制対応と社内ルールの整備、使いやすさとノウハウの不足、そして情報漏洩などのセキュリティ懸念の3点を挙げる。

「安心して社内で利用できるルール整備を行ったうえで、Alli LLM App Marketを導入することで、利用しやすく、セキュアに生成AIを活用できます。また、自社のセキュリティポリシーに適合させた環境を実現することで秘匿性の高い業務情報も利用可能です」（小林氏）

小林氏は、効果的な全社展開方法の事例として、同社が支援を行った従業員5,000人規模のITサービス企業でのAlli LLM App Marketの導入事例について解説した。この企業では、多くの社員が生成AIに触れて、便利さを実感してもらうことで利用の拡大を図る方針を立て、全社展開を加速するために、部門でのトライアルを行わずに、まず全社公開を行ったことが特徴的だ。社内ポータルサイトを通じたAlli LLM App Marketへの容易なアクセス確保や、親しみやすい名称の採用により、心理的なハードルを下げることに成功したという。

セキュリティ面では、シングルサインオン認証の導入や、全社活用と特定業務用に環境を分離する���ど、社内ポリシーに応じた柔軟な対応を実現。日立ソリューションズでは生成AI導入の豊富な実績とノウハウをもとに、SaaSから物理サーバー環境までさまざまな導入形態に対応可能だ。

「今後は生成AIを業務システムやプロセスに組み込み、業務全体を効率化していく流れが加速するでしょう」と小林氏は展望を語る。たとえば問い合わせ対応業務では、受付後の手間のかかる回答作成、進捗管理などを含めた一連のプロセスをデジタル化し、そこに生成AIを組み込むことで、より効率的な業務遂行が可能になるという。

日立ソリューションズでは、今後、こうした業務全体の効率化を実現するソリューションの提供を進めていく。

関連リンク 生成AI（Generative AI）とは？生成AIサービスをビジネスで活用する導入支援 https://www.hitachi-solutions.co.jp/products/pickup/generative-ai/

企業向け生成AI利用環境を提供する「Alli LLM App Market」 https://www.hitachi-solutions.co.jp/allganize/

日立ソリューションズが提供する業務全体の効率化を実現するソリューション （活文 業務プロセスデジタル化ソリューション） https://www.hitachi-solutions.co.jp/katsubun/bpds/

[PR]提供：日立ソリューションズ

(セキュリティと利便性を両立させた「生成AI活用環境」の実現方法──日立ソリューションズとAllganize Japanが示す実践的アプローチ | TECH+（テックプラス）から)

[noticiassincensura]

Amazon’s AI Race Mystery: $8 Billion Invested and No Product to Show

The company has just doubled its investment in Anthropic but has yet to offer any tangible AI solutions.

All Big Tech companies have something to show in the AI space — except Amazon, which remains low-key for now. The company has yet to announce any groundbreaking developments in AI and seems unlikely to do so until 2025. However, it is pouring immense resources into this sector, recently making another substantial investment. The concerning part is that, so far, this expenditure hasn’t materialized into a visible product.

Another $4 Billion for Amazon

Amazon has announced a $4 billion investment in Anthropic, OpenAI’s rival and creator of the Claude chatbot. This mirrors the $4 billion investment Amazon made in the same company in March 2024, reinforcing its position as a significant player in one of the sector’s key players.

Another AI Startup Giant

Rumors about a potential investment round for Anthropic had been circulating for weeks. Both OpenAI and xAI recently completed massive funding rounds, increasing their market valuations. With this move, Amazon positions Anthropic as a key player in the field. According to Crunchbase, Anthropic has raised $13.7 billion, with $8 billion of that coming from Amazon.

Training on AWS

As part of the agreement, Anthropic will primarily train its generative AI models on Amazon Web Services (AWS). This is similar to the Microsoft-OpenAI deal, where OpenAI heavily uses Azure services instead of competitors.

Moving Away from NVIDIA

Anthropic will leverage Amazon’s Trainium2 chips for training and Inferentia chips for running its AI models. Previously, the startup relied heavily on NVIDIA’s chips for training. With this new agreement, Anthropic commits to focusing its training and inference processes on Amazon’s solutions.

Future Chips

Anthropic will also collaborate with Amazon to develop specialized AI chips. Engineers from both organizations will work with Annapurna Labs, Amazon’s division for chip development. The goal is to create future generations of the Trainium accelerator, designed for more efficient and powerful AI model training.

What About Amazon’s AI?

Amazon’s significant investment in Anthropic hasn’t yet translated into a visible product. This contrasts with Microsoft’s investment in OpenAI, which quickly led to its Copilot family of solutions, with ChatGPT as a cornerstone, being integrated across Microsoft’s ecosystem. Amazon, however, has yet to release a chatbot or generative AI services for end users, though it has launched some projects, such as Amazon Q, an AI chatbot for businesses.

Alexa with More AI on the Horizon

Amazon’s main AI initiative seems to be a relaunch of Alexa. Its voice assistant, which powers devices like Amazon Echo, may be revamped as “Remarkable Alexa,” featuring much more advanced conversational capabilities. This version could potentially be subscription-based, similar to ChatGPT Plus. However, it’s unclear if it will be based on Amazon’s own LLM. Recent reports suggest that Amazon might build this advanced Alexa on Claude, Anthropic’s chatbot.

Metis and Olympus in the Background

In June, reports revealed Amazon has been developing its own LLM, called Olympus, aimed at competing with models like GPT-4, Gemini, or Claude 3.5 Sonnet. This AI model could be integrated into Alexa and also offered through a web-based service named Metis, essentially Amazon’s version of ChatGPT.

But Questions Remain

These developments are yet to materialize, raising doubts about Amazon’s relevance in the AI sector. The company seems to have missed the generative AI train but might be waiting to launch a well-polished product. Apple, which has also been slow with its Apple Intelligence features, is another Big Tech company that has disappointed in this space. Time will tell if Amazon follows suit or makes a strong entry.

[drmikewatts]

Weekly Review 8 November 2024

Some interesting links that I Tweeted about in the last week (I also post these on Mastodon, Threads, Newsmast, and Bluesky):

AI that build better AI, without human involvement or intervention, is something we need to be very careful about: https://arstechnica.com/ai/2024/10/the-quest-to-use-ai-to-build-better-ai/

Honestly, he's not wrong about AI being hyped. And I agree that in time it will become useful, once the hype has died down: https://www.tomshardware.com/tech-industry/artificial-intelligence/linus-torvalds-reckons-ai-is-90-percent-marketing-and-10-percent-reality

Web search is another area where AI is taking over: https://www.bigdatawire.com/2024/11/01/openai-and-google-clash-in-the-evolution-of-ai-powered-search/

AI services is having a small but measurable impact on Microsoft's profitability: https://arstechnica.com/gadgets/2024/10/microsoft-reports-big-profits-amid-massive-ai-investments/

You don't need GPU to run AI, it can be done in CPU: https://www.theregister.com/2024/10/29/cpu_gen_ai_gpu/

How AI is affecting jobs and the workplace: https://www.datasciencecentral.com/the-impact-of-ai-powered-automation-on-workforce-dynamics-and-job-roles/

If the training data isn't open, then the AI isn't open: https://www.bigdatawire.com/2024/10/28/osi-open-ai-definition-stops-short-of-requiring-open-data/

Another way AI is affecting the climate-AI run in data centers, which use a lot of concrete in their construction, and concrete production releases carbon: https://spectrum.ieee.org/green-concrete

A point-by-point overview of ChatGPT: https://www.techrepublic.com/article/gpt-4-cheat-sheet/

Generative AI is now being rolled-out to Gmail: https://www.theverge.com/2024/10/28/24282103/gmail-help-me-write-email-web-ai-gemini

Here the AI is helping programmers be more productive, rather than replacing them. But given the known security issues with AI-generated code, is it too much to have 25% generated by AI? https://arstechnica.com/ai/2024/10/google-ceo-says-over-25-of-new-google-code-is-generated-by-ai/

Generative AI comes with a lot of legal risks: https://www.informationweek.com/machine-learning-ai/the-intellectual-property-risks-of-genai

Five things that Generative AI is expected to impact in 2025: https://www.techrepublic.com/article/generative-ai-trends-2025/

Microsoft is focusing on running AI inferencing in Azure rather than training: https://www.theregister.com/2024/10/31/microsoft_q1_fy_2025/

A swarm of cooperating agents might be the way to truly powerful AI: https://www.computerworld.com/article/3594235/agentic-ai-swarms-are-headed-your-way.html

An overview of AI in healthcare: https://www.datasciencecentral.com/how-ai-is-shaping-the-future-of-the-healthcare-industry/

You could achieve general AI with a billion people using abacuses. That doesn't mean it's feasible: https://futurism.com/sam-altman-agi-achievable-current-hardware

Am I being cynical in thinking that an AI powered web search engine is going to hallucinate web sites? https://www.stuff.co.nz/world-news/360472566/openai-adds-search-chatgpt-challenging-google

The current tools an AI developer needs to be familiar with: https://www.informationweek.com/machine-learning-ai/the-essential-tools-every-ai-developer-needs

Good clean data is essential for training AI. Here are ten Python commands that help clean data: https://www.kdnuggets.com/10-useful-python-one-liners-for-data-cleaning

Combining AI with Google maps: https://www.theverge.com/2024/10/31/24283970/google-maps-gemini-ai-answer-questions

This is the best use of AI in journalism-using it to support their work by transcribing recordings, rather than trying to replace the reporters entirely: https://arstechnica.com/ai/2024/10/the-new-york-times-shows-how-ai-can-aid-reporters-without-replacing-them/

If you're training your AI with other people's work, you really should know what plagiarism is: https://techcrunch.com/2024/10/30/perplexitys-ceo-punts-on-defining-plagiarism/

Giving instructions in hexadecimal can defeat AI guardrails, in this case tricking ChatGPT into writing exploit code: https://www.theregister.com/2024/10/29/chatgpt_hex_encoded_jailbreak/

[7ooo-ru]

OpenAI готовит флагманскую ИИ-модель: в 100 раз мощнее GPT-4

По предварительным данным, выход состоится уже совсем скоро. Однако есть нюанс.

Компания OpenAI готовится запустить свою следующую ИИ-модель. Она получила название Orion. Планы разработчиков раскрыло издание The Verge, которое ссылается на несколько источников.

Ожидается, что выход Orion будет отличаться от прошлых релизов. Если GPT-4o и o1 сразу представили пользователям ChatGPT, то сейчас первыми доступ получат некоторые компании. Предполагается, что сначала нейросетью смогут воспользоваться организации, с которым OpenAI тесно сотрудничает. Благодаря этому компании смогут создать собственные продукты и функции.

Также есть информация, что корпорация Microsoft, важный партнер OpenAI, может добавить Orion на свою платформу Azure уже в ноябре. Однако основной релиз запланирован на декабрь. Другие подробности пока что не раскрываются.

OpenAI планирует сделать Orion преемником GPT-4. Сейчас нет информации о том, будут ли новую модель называть GPT-5. Ранее глава компании Сэм Альтман заявил, что Orion может оказаться в 100 раз более мощной, чем GPT-4. При этом новая нейросеть будет отделена от модели рассуждений o1. В будущем разработчики собираются объединить свои большие языковые модели, чтобы взять преимущества каждой версии.

Подробнее https://7ooo.ru/group/2024/10/25/970-openai-gotovit-flagmanskuyu-ii-model-v-100-raz-moschnee-gpt-4-grss-350179472.html

[yourepfan]

How Apple’s AI is Years Behind Competitors: A Deep Dive

In the rapidly evolving world of artificial intelligence (AI), some tech giants lead the charge while others struggle to keep up. While Apple is widely regarded for its innovation in hardware and design, it is lagging behind in one crucial area—AI. Companies like Google, Microsoft, and OpenAI have surged ahead, leaving Apple grappling with a future increasingly defined by artificial intelligence. In this article, we'll explore the reasons Apple’s AI strategy is years behind its competitors and what that could mean for the future of the tech giant.

1. Siri's Stagnation

Apple was once a pioneer in AI-driven voice assistants with the release of Siri in 2011. Initially seen as a breakthrough, Siri has since failed to keep pace with rivals like Amazon Alexa, Google Assistant, and even newer systems like ChatGPT from OpenAI. While Alexa and Google Assistant have become household names known for their deep integrations, better conversational capabilities, and broader functionality, Siri remains comparatively rigid and lacks the same level of contextual understanding and adaptability.

Key Problems with Siri:

Limited conversational depth: Siri often fails to engage in multi-turn conversations or handle complex queries.

Less integration with third-party apps: While Google Assistant can interact seamlessly with thousands of third-party services, Siri is still limited in scope.

Slow learning curve: Siri's ability to improve based on user interactions seems minimal compared to the fast-learning AI models seen in other assistants.

2. Lack of AI-Focused Hardware and Infrastructure

Apple has always excelled in creating beautifully designed and highly functional hardware, but its AI capabilities are not well-supported by its hardware ecosystem. Google's Tensor Processing Units (TPUs) and Nvidia’s GPUs, for example, are pushing AI computations forward at an unprecedented pace. Microsoft’s cloud infrastructure, built through Azure, supports AI services that cater to large-scale enterprise needs.

In contrast, Apple’s hardware is not as well-suited for cutting-edge AI development. While the company has made strides with its in-house chips like the A-series and M-series processors, these are more geared towards general computing and efficiency rather than AI-specific tasks. Apple lacks the kind of AI-focused infrastructure seen in competitors, putting it behind in areas like machine learning model training and large-scale AI deployment.

3. Lack of Open AI Development

OpenAI, Google, and Microsoft are creating a significant impact by democratizing access to AI technology. For example, OpenAI’s GPT models are open to developers and businesses via API, enabling others to build on top of their technology. Google’s AI research is also openly available, providing valuable contributions to the broader scientific community.

Apple’s approach, in contrast, is much more closed. The company has always prioritized privacy and security, which is commendable, but this philosophy has also led to a restrictive AI development environment. Apple doesn’t offer the same level of open tools, frameworks, or APIs for AI development, slowing innovation and limiting the broader tech ecosystem's ability to build on its AI technologies.

4. Apple’s Privacy-First Approach is a Double-Edged Sword

Apple's commitment to user privacy is one of its defining principles. This focus on privacy makes Apple's AI solutions, such as Siri, more cautious in terms of data collection and usage compared to its competitors. However, this also limits the company's ability to use data to train advanced AI models. Competitors like Google have access to enormous datasets, allowing them to develop AI systems that can learn from billions of interactions and provide personalized experiences at scale.

For instance, Google Assistant uses data from search queries, emails, and even location to provide highly tailored responses, while Siri’s functionality remains relatively basic. Apple's privacy-first approach, while important, puts constraints on its ability to innovate quickly in the AI space, where data is essential for improving performance and capabilities.

5. Delayed AI Integration Across Products

Another major factor is Apple’s sluggish integration of AI into its core products and services. Companies like Google and Microsoft are embedding AI into nearly every product, from search engines and web browsers to enterprise-level cloud services. Microsoft, for example, has incorporated AI into its Office Suite (e.g., Excel and Word) and is leveraging OpenAI’s GPT models across its entire ecosystem.

Apple, on the other hand, has been slow to integrate AI meaningfully beyond a few features in Photos, Siri, and iOS predictive text. While its products benefit from machine learning in terms of performance, battery life, and camera features, Apple is not innovating at the same scale when it comes to leveraging AI across its ecosystem.

6. Underwhelming AI Acquisitions

While Apple has made numerous AI-related acquisitions over the years, it hasn't translated them into groundbreaking consumer-facing technologies. Companies like Google have used acquisitions to rapidly advance their AI capabilities, but Apple’s acquisitions—such as Turi (a machine learning company) and Xnor.ai (edge-based AI)—haven't resulted in significant improvements in its core products. Meanwhile, competitors like Microsoft have made strategic acquisitions like OpenAI, giving them an enormous advantage in large language models and generative AI.

7. Competitors Are Moving Faster

The world of AI moves at breakneck speed, and Apple has not matched the urgency of its competitors. OpenAI's iterative advancements with GPT models, Google’s continuous improvements in areas like search and cloud AI, and Microsoft’s aggressive AI-driven strategies in enterprise software are setting the pace. Apple, meanwhile, continues to focus on refining its user experience and hardware design, which, while valuable, doesn't place them at the forefront of the AI revolution.

The Road Ahead: Can Apple Catch Up?

Despite its slow progress in AI, Apple still has considerable resources and brand loyalty to leverage. The company's strength lies in its ability to create seamless hardware-software experiences, and there’s potential for Apple to use AI in innovative ways within this ecosystem. For example, integrating AI-driven health features in its wearables, or making Siri more contextually aware and intelligent, could give Apple a unique edge.

However, to catch up, Apple will need to significantly ramp up its AI research and development, consider opening up its AI platforms to developers, and potentially ease some of its privacy constraints in a responsible manner. Without a bold move, Apple risks becoming an AI follower rather than a leader in the next wave of technological innovation.

In conclusion, Apple has built its empire on revolutionary design, seamless user experience, and premium hardware, but as AI becomes the cornerstone of future technology, its reluctance or inability to lead in AI innovation puts it in a precarious position. While the company is far from out of the game, it must shift gears if it hopes to keep pace in the AI arms race.

Contact Us

[tsrtimes]

Empower Your Digital Transformation with Microsoft Azure Cloud Service

Today, cloud computing applications and platforms are promptly growing across various industries, allowing businesses to become more efficient, effective, and competitive. In fact, these days, over 77% of businesses have some part of their computing infrastructure in the cloud.

Although, there are various cloud computing platforms available, one of the few platforms that lead the cloud computing industry is Microsoft Azure Cloud. Although, Amazon Web Services (AWS) is a leading giant in the public cloud market, Azure is the most rapidly-growing and second-largest in the world of computing.

What is Microsoft Azure?

Azure is a cloud computing service provided by Microsoft. There are more than six hundred services that come under the Azure umbrella. In simple terms, it is a web-based platform used for building, testing, managing and deploying applications and services.

About 80% of Fortune 500 Companies are using Azure for their cloud computing requirements.

Azure supports a multitude of programming languages, including Node JS, Java and C#.

Another interesting fact about Azure is that it has nearly 42 data centers around the globe, which is the maximum number of data centers for any cloud platform.

A broad range of Microsoft’s Software-as-a-Service (SaaS), Platform-as-a-Service (PaaS) and Infrastructure-as-a-Service (IaaS) products are hosted on Azure. To understand these major cloud computing service models in detail, check out our other blog.

Azure provides three key aspects of functionality: Virtual Machine, app services and cloud services.

Virtual Machines

The virtual machines by Azure are one of many types of scalable, on-demand computing resources. An Azure virtual machine provides you with the flexibility of virtualization without the need of buying and maintaining the physical hardware that runs it.

App Services

Azure App Services is a service based on HTTP which is used for hosting web applications, mobile back ends, and REST APIs. It can be developed in your favourite language, be it JAVA, .NET, .NET Core, Node.js, Ruby, PHP or Python. Applications smoothly run and scale on both Windows and Linux-based environments.

Cloud Services

Azure Cloud Services is a form of Platform-as-a-Service. Similar to Azure App Service, the technology is crafted to support applications that are reliable, scalable, and reasonable to operate. Like App Services are hosted on virtual machines, so do Azure Cloud Services.

Various Azure services and how it works

Azure offers over 200 services, divided across 18 categories. These categories constitute computing, storage, networking, IoT, mobile, migration, containers, analytics, artificial intelligence and other machine learning, management tools, integration, security, developer tools, databases, security, DevOps, media identity and web services. Below, we have broken down some of these important Azure services based on their category:

Computer services

Azure Cloud Service: You can create measurable applications within the cloud by using this service. It offers instant access to the latest services and technologies required in the enterprise, enabling Azure cloud engineers to execute complex solutions seamlessly.

Virtual Machines: They offer Infrastructure-as-a-Service and can be used in diverse ways. When there is a need for complete control over an operating system and environment, VMS are a suitable choice. With this service, you can create a virtual machine in Linux, Windows or any other configuration in seconds.

Service Fabric: It is a Platform-as-a-Service which is designed to facilitate the development, deployment and management of highly customizable and scalable applications for the Microsoft Azure cloud platform. It simplifies the process of developing a micro service.

Functions: It enables you to build applications in any programming language. When you’re simply interested in the code that runs your service and not the underlying platform or infrastructure, functions are great.

Networking

Azure CDN: It helps store and access data on varied content locations and servers. Using Azure CDN (Content Delivery Network), you can transfer content to any person around the world.

Express Route: This service allows users to connect their on-premise network to the Microsoft Cloud or any other services using a private connection.  ExpressRoute offers more reliability, consistent latencies, and faster speed than usual connections on the internet.

Virtual Network: It is a logical representation of the network in the cloud. So, by building an Azure Virtual Network, you can decide your range of private IP addresses. This service enables users to have any of the Azure services communicate with each other securely and privately.

Azure DNS: Name resolution is provided by Azure DNS, a hosting service for DNS domains that makes use of the Microsoft Azure infrastructure. You can manage your DNS records by utilising the same login information, APIs, tools, and pricing as your other Azure services if you host your domains in Azure.

Storage

Disk Storage: In Azure, VM uses discs as a storage medium for an operating system, programs, and data. A Windows operating system disc plus a temporary disc are the minimum numbers of discs present in any virtual machine.

File Storage: The main usage of Azure file storage is to create a shared drive between two servers or users. We’ll use Azure file storage in that situation. It is possible to access this managed file storage service via the server message block (SMB) protocol.

Blob Storage: Azure blob storage is essential to the overall Microsoft Azure platform since many Azure services will store and act on data that is stored in a storage account inside the blob storage. And each blob must be kept in its own container.

Benefits of using Azure

Application development: Any web application can be created in Azure.

Testing: After the successful development of the application on the platform, it can be easily tested.

Application hosting: After the testing, you can host the application with the help of Azure.

Create virtual machines: Using Azure, virtual machines can be created in any configuration.

Integrate and sync features: Azure enables you to combine and sync directories and virtual devices

Collect and store metrics: Azure allows you to collect and store metrics, enabling you to identify what works.

Virtual hard drives: As they are extensions of virtual machines, they offer a massive amount of data storage.

Bottom line

With over 200 services and countless benefits, Microsoft Azure Cloud is certainly the most rapidly-growing cloud platform being used by organizations. Incessant innovation from Microsoft allows businesses to respond quickly to unexpected changes and new opportunities.

So, are you planning to migrate your organization’s data and workload to the cloud? At CloudScaler, get instant access to the best services and technologies from the ground up, supported by a team of experts that keep you one step ahead in the competition.

[jcmarchi]

Generative AI, innovation, creativity & what the future might hold - CyberTalk

New Post has been published on https://thedigitalinsider.com/generative-ai-innovation-creativity-what-the-future-might-hold-cybertalk/

Generative AI, innovation, creativity & what the future might hold - CyberTalk

Stephen M. Walker II is CEO and Co-founder of Klu, an LLM App Platform. Prior to founding Klu, Stephen held product leadership roles Productboard, Amazon, and Capital One.

Are you excited about empowering organizations to leverage AI for innovative endeavors? So is Stephen M. Walker II, CEO and Co-Founder of the company Klu, whose cutting-edge LLM platform empowers users to customize generative AI systems in accordance with unique organizational needs, resulting in transformative opportunities and potential.

In this interview, Stephen not only discusses his innovative vertical SaaS platform, but also addresses artificial intelligence, generative AI, innovation, creativity and culture more broadly. Want to see where generative AI is headed? Get perspectives that can inform your viewpoint, and help you pave the way for a successful 2024. Stay current. Keep reading.

Please share a bit about the Klu story:

We started Klu after seeing how capable the early versions of OpenAI’s GPT-3 were when it came to common busy-work tasks related to HR and project management. We began building a vertical SaaS product, but needed tools to launch new AI-powered features, experiment with them, track changes, and optimize the functionality as new models became available. Today, Klu is actually our internal tools turned into an app platform for anyone building their own generative features.

What kinds of challenges can Klu help solve for users?

Building an AI-powered feature that connects to an API is pretty easy, but maintaining that over time and understanding what’s working for your users takes months of extra functionality to build out. We make it possible for our users to build their own version of ChatGPT, built on their internal documents or data, in minutes.

What is your vision for the company?

The founding insight that we have is that there’s a lot of busy work that happens in companies and software today. I believe that over the next few years, you will see each company form AI teams, responsible for the internal and external features that automate this busy work away.

I’ll give you a good example for managers: Today, if you’re a senior manager or director, you likely have two layers of employees. During performance management cycles, you have to read feedback for each employee and piece together their strengths and areas for improvement. What if, instead, you received a briefing for each employee with these already synthesized and direct quotes from their peers? Now think about all of the other tasks in business that take several hours and that most people dread. We are building the tools for every company to easily solve this and bring AI into their organization.

Please share a bit about the technology behind the product:

In many ways, Klu is not that different from most other modern digital products. We’re built on cloud providers, use open source frameworks like Nextjs for our app, and have a mix of Typescript and Python services. But with AI, what’s unique is the need to lower latency, manage vector data, and connect to different AI models for different tasks. We built on Supabase using Pgvector to build our own vector storage solution. We support all major LLM providers, but we partnered with Microsoft Azure to build a global network of embedding models (Ada) and generative models (GPT-4), and use Cloudflare edge workers to deliver the fastest experience.

What innovative features or approaches have you introduced to improve user experiences/address industry challenges?

One of the biggest challenges in building AI apps is managing changes to your LLM prompts over time. The smallest changes might break for some users or introduce new and problematic edge cases. We’ve created a system similar to Git in order to track version changes, and we use proprietary AI models to review the changes and alert our customers if they’re making breaking changes. This concept isn’t novel for traditional developers, but I believe we’re the first to bring these concepts to AI engineers.

How does Klu strive to keep LLMs secure?

Cyber security is paramount at Klu. From day one, we created our policies and system monitoring for SOC2 auditors. It’s crucial for us to be a trusted partner for our customers, but it’s also top of mind for many enterprise customers. We also have a data privacy agreement with Azure, which allows us to offer GDPR-compliant versions of the OpenAI models to our customers. And finally, we offer customers the ability to redact PII from prompts so that this data is never sent to third-party models.

Internally we have pentest hackathons to understand where things break and to proactively understand potential threats. We use classic tools like Metasploit and Nmap, but the most interesting results have been finding ways to mitigate unintentional denial of service attacks. We proactively test what happens when we hit endpoints with hundreds of parallel requests per second.

What are your perspectives on the future of LLMs (predictions for 2024)?

This (2024) will be the year for multi-modal frontier models. A frontier model is just a foundational model that is leading the state of the art for what is possible. OpenAI will roll out GPT-4 Vision API access later this year and we anticipate this exploding in usage next year, along with competitive offerings from other leading AI labs. If you want to preview what will be possible, ChatGPT Pro and Enterprise customers have access to this feature in the app today.

Early this year, I heard leaders worried about hallucinations, privacy, and cost. At Klu and across the LLM industry, we found solutions for this and we continue to see a trend of LLMs becoming cheaper and more capable each year. I always talk to our customers about not letting these stop your innovation today. Start small, and find the value you can bring to your customers. Find out if you have hallucination issues, and if you do, work on prompt engineering, retrieval, and fine-tuning with your data to reduce this. You can test these new innovations with engaged customers that are ok with beta features, but will greatly benefit from what you are offering them. Once you have found market fit, you have many options for improving privacy and reducing costs at scale – but I would not worry about that in the beginning, it’s premature optimization.

LLMs introduce a new capability into the product portfolio, but it’s also an additional system to manage, monitor, and secure. Unlike other software in your portfolio, LLMs are not deterministic, and this is a mindset shift for everyone. The most important thing for CSOs is to have a strategy for enabling their organization’s innovation. Just like any other software system, we are starting to see the equivalent of buffer exploits, and expect that these systems will need to be monitored and secured if connected to data that is more important than help documentation.

Your thoughts on LLMs, AI and creativity?

Personally, I’ve had so much fun with GenAI, including image, video, and audio models. I think the best way to think about this is that the models are better than the average person. For me, I’m below average at drawing or creating animations, but I’m above average when it comes to writing. This means I can have creative ideas for an image, the model will bring these to life in seconds, and I am very impressed. But for writing, I’m often frustrated with the boring ideas, although it helps me find blind spots in my overall narrative. The reason for this is that LLMs are just bundles of math finding the most probable answer to the prompt. Human creativity —from the arts, to business, to science— typically comes from the novel combinations of ideas, something that is very difficult for LLMs to do today. I believe the best way to think about this is that the employees who adopt AI will be more productive and creative— the LLM removes their potential weaknesses, and works like a sparring partner when brainstorming.

You and Sam Altman agree on the idea of rethinking the global economy. Say more?

Generative AI greatly changes worker productivity, including the full automation of many tasks that you would typically hire more people to handle as a business scales. The easiest way to think about this is to look at what tasks or jobs a company currently outsources to agencies or vendors, especially ones in developing nations where skill requirements and costs are lower. Over this coming decade you will see work that used to be outsourced to global labor markets move to AI and move under the supervision of employees at an organization’s HQ.

As the models improve, workers will become more productive, meaning that businesses will need fewer employees performing the same tasks. Solo entrepreneurs and small businesses have the most to gain from these technologies, as they will enable them to stay smaller and leaner for longer, while still growing revenue. For large, white-collar organizations, the idea of measuring management impact by the number of employees under a manager’s span of control will quickly become outdated.

While I remain optimistic about these changes and the new opportunities that generative AI will unlock, it does represent a large change to the global economy. Klu met with UK officials last week to discuss AI Safety and I believe the countries investing in education, immigration, and infrastructure policy today will be best suited to contend with these coming changes. This won’t happen overnight, but if we face these changes head on, we can help transition the economy smoothly.

Is there anything else that you would like to share with the CyberTalk.org audience?

Expect to see more security news regarding LLMs. These systems are like any other software and I anticipate both poorly built software and bad actors who want to exploit these systems. The two exploits that I track closely are very similar to buffer overflows. One enables an attacker to potentially bypass and hijack that prompt sent to an LLM, the other bypasses the model’s alignment tuning, which prevents it from answering questions like, “how can I build a bomb?” We’ve also seen projects like GPT4All leak API keys to give people free access to paid LLM APIs. These leaks typically come from the keys being stored in the front-end or local cache, which is a security risk completely unrelated to AI or LLMs.

[enterprisewired]

OpenAI Secures $4 Billion Credit Line Amid Rapid Growth and Expansion Plans

[Source – engadget.com]

OpenAI has secured a $4 billion revolving credit line, bringing its total liquidity to more than $10 billion, CNBC has learned. The credit line, provided by a group of major financial institutions including JPMorgan Chase, Citi, Goldman Sachs, and Morgan Stanley, comes as OpenAI continues its aggressive push into AI research, infrastructure development, and talent acquisition. This financing follows a recent funding round that valued the company at $157 billion.

Strategic Financial Flexibility for Expansion

The $4 billion credit line includes an option to increase it by an additional $2 billion, giving OpenAI substantial financial flexibility. The loan, which is unsecured, can be accessed over three years, with an interest rate of approximately 6%, tied to the Secured Overnight Financing Rate (SOFR).

In a blog post, OpenAI emphasized the importance of this liquidity, stating, “This gives us the flexibility to invest in new initiatives and operate with full agility as we scale.” The funds will primarily be used to support research, expand infrastructure, and attract top talent as the company seeks to maintain its leadership position in the fast-evolving AI sector.

Record-setting growth and Significant Investments

OpenAI’s meteoric rise began with the launch of ChatGPT in late 2022, bringing generative AI into the mainstream and attracting tens of billions of dollars in investments. The company’s rapid growth has led to a surge in revenue, with $300 million generated last month alone—a 1,700% increase since early 2023. OpenAI projects $11.6 billion in sales for 2025, up from an expected $3.7 billion in 2024.

However, the company’s growth comes at a cost. OpenAI anticipates a loss of $5 billion this year, largely due to high expenses tied to purchasing Nvidia graphics processing units needed to train its large language models. Despite these challenges, OpenAI’s partnership with Microsoft, which has invested billions, has been key in bolstering its Azure cloud business.

Leadership Changes and Plans for Restructuring

OpenAI has faced internal challenges as well, including the departure of key executives like CTO Mira Murati and research chief Bob McGrew. Amid these transitions, the company’s board is exploring restructuring options, potentially moving OpenAI from its current model to a more traditional for-profit structure. CEO Sam Altman recently denied rumors of receiving a large equity stake in the company, while CFO Sarah Friar discussed the company’s long-term aspirations and capital strategies in a CNBC interview.

OpenAI is exploring diverse financing options, including public and debt markets, as it aims to position itself as a sustainable, long-term player in the AI industry. The company’s board continues to discuss whether compensating key executives with equity would benefit its mission, although no decisions have been made.

[govindhtech]

Boost the development of AI apps with Cloud Modernization

Cloud Modernization

A new era of intelligent applications that can comprehend natural language, produce material that is human-like, and enhance human abilities has begun with the development of generative AI. But as businesses from all sectors start to see how AI may completely transform their operations, they frequently forget to update their on-premises application architecture, which is an essential first step.

Cloud migration

Cloud migration is significantly more advantageous than on-premises options if your company wants to use AI to improve customer experiences and spur growth. Numerous early adopters, including TomTom and H&R Block, have emphasized that their decision to start updating their app architecture on Azure was what prepared them for success in the AI era.

Further information to connect the dots was provided by a commissioned study by IDC titled “Exploring the Benefits of Cloud Migration and Cloud Modernization for the Development of Intelligent Applications,” which was based on interviews with 900 IT leaders globally regarding their experiences moving apps to the cloud. They’ll go over a few of the key points in this article.

Modernise or lag behind: The necessity of cloud migration driven by AI

Let’s say what is obvious: Artificial Intelligence is a potent technology that can write code, produce content, and even develop whole apps. The swift progress in generative AI technologies, such OpenAI’s GPT-4 has revolutionized the way businesses function and engage with their clientele.

However, generative AI models such as those that drive ChatGPT or image-generating software are voracious consumers of data. To achieve their disruptive potential, they need access to enormous datasets, flexible scaling, and immense computing resources. The computation and data needs of contemporary AI workloads are simply too much for on-premises legacy systems and compartmentalized data stores to handle.

Cloud Modernization systems, which are entirely managed by the provider, offer the reliable infrastructure and storage options required to handle AI workloads. Because of its nearly infinite scalability, apps can adapt to changing demand and continue to operate at a high level.

The main finding of the IDC survey was that businesses were mostly driven to move their applications to the cloud by a variety of benefits, such as enhanced security and privacy of data, easier integration of cloud-based services, and lower costs. Furthermore, companies can swiftly test, refine, and implement AI models because to the cloud’s intrinsic agility, which spurs innovation.

With its most recent version, the.NET framework is ready to use AI in cloud settings. Developers can use libraries like OpenAI, Qdrant, and Milvus as well as tools like the Semantic Kernel to include AI capabilities into their apps. Applications may be delivered to the cloud with excellent performance and scalability thanks to the integration with.

NET Aspire. H&R Block’s AI Tax Assistant, for instance, shows how companies may build scalable, AI-driven solutions to improve user experiences and operational efficiency. It was developed using.NET and Azure OpenAI. You may expedite development and boost the adoption of AI in all areas of your company operations by integrating. NET into your cloud migration plan.

Utilising cloud-optimized old on-premises programmes through migration and restructuring allows for the seamless scaling of computation, enormous data repositories, and AI services. This can help your business fully incorporate generative AI into all aspects of its data pipelines and intelligent systems, in addition to allowing it to develop generative AI apps.

Reach your AI goals faster in the cloud

The ambition of an organisation to use generative AI and the realisation of its full value through cloud migration are strongly correlated, according to a recent IDC study. Let’s dissect a few important factors:

Data accessibility: Consolidating and accessing data from several sources is made easier by cloud environments, giving AI models the knowledge they require for training and improvement.

Computational power: Elastic computing resources in the cloud may be flexibly distributed to fulfil complicated AI algorithm needs, resulting in optimal performance and cost effectiveness.

Collaboration: Data scientists, developers, and business stakeholders may work together more easily thanks to cloud-based tools, which speeds up the creation and application of AI.

Cloud migration speeds up innovation overall in addition to enabling generative AI. Cloud platforms offer an abundance of ready-to-use services, such as serverless computing, machine learning, and the Internet of Things, that enable businesses to quickly develop and integrate new intelligent features into their apps.

Adopt cloud-based AI to beat the competition

Gaining a competitive edge is the driving force behind the urgent need to migrate and modernise applications it’s not simply about keeping up with the times. Companies who use AI and the cloud are better positioned to:

Draw in elite talent Companies with state-of-the-art tech stacks attract the best data scientists and developers.

Adjust to shifting circumstances: Because of the cloud’s elasticity, organisations may quickly adapt to changing client wants or market conditions.

Accelerate the increase of revenue: Applications driven by AI have the potential to provide new revenue streams and improve consumer satisfaction.

Embrace AI-powered creativity by updating your cloud

Cloud migration needs to be more than just moving and lifting apps if it is to stay competitive. The key to unlocking new levels of agility, scalability, and innovation in applications is Cloud Modernization through rearchitecting and optimizing them for the cloud. Your company can: by updating to cloud-native architectures, your apps can:

Boost performance: Incorporate intelligent automation, chatbots, and personalised recommendations all enabled by AI into your current applications.

Boost output: To maximise the scalability, responsiveness, and speed of your applications, take advantage of cloud-native technology.

Cut expenses: By only paying for the resources you use, you can do away with the expensive on-premises infrastructure.

According to the IDC poll, most respondents decided to move their apps to the Cloud Modernization because it allowed them to develop innovative applications and quickly realize a variety of business benefits.

Boost the development of your intelligent apps with a cloud-powered AI

In the age of generative AI, moving and updating apps to the cloud is not a choice, but a requirement. Businesses that jump on this change quickly will be in a good position to take advantage of intelligent apps’ full potential, which will spur innovation, operational effectiveness, and consumer engagement.

The combination of generative AI and cloud computing is giving organizations previously unheard-of options to rethink their approaches and achieve steady growth in a cutthroat market.

Businesses may make well-informed decisions on their cloud migration and Cloud Modernization journeys by appreciating the benefits and measuring the urgency, which will help them stay at the forefront of technical advancement and commercial relevance.

Read more on Govindhtech.com

[uwteam]

24 maja 2024

◢ #unknownews ◣

Zapraszam do lektury dzisiejszego zestawienia. Za tydzień nie planuję wysyłki newslettera, ponieważ będę na urlopie.

Sponsorem tego wydania newslettera jest podcast Patoarchitekci. Chłopaki opublikowali kalendarz szkoleń na ten rok – praktyczne warsztaty online z architektury systemowej, DevOps, automatyzacji rozwiązań bazujacymi na chmurze Azure jak i na CNCF. Kursy prowadzone sa w formie warsztatów online na żywo. Zobacz, co mają w ofercie.

1) Inżynieria wsteczna aplikacji Electron w poszukiwaniu API https://danaepp.com/reverse-engineering-electron-apps-to-discover-apis INFO: Czy kiedykolwiek zastanawiałeś się, co dzieje się pod maską aplikacji desktopowej, której używasz? Autor przedstawia techniki reverse engineeringu na przykładzie konkretnej aplikacji. Dzięki temu udaje mu się dobrać do endpointów API, z których ona korzysta.

2) Czy to koniec internetu jaki znamy? - zatrucie źródeł wiedzy (film, 21m) https://youtu.be/EXk0IYtSpes?si=kxjf0KkpmvLZQ9-p INFO: Czy można zatruć źródło wiedzy, z którego uczą się LLM-y? Czy zatrucie LLM-ów wpłynie bezpośrednio na cały Internet? Jaki związek z tym wszystkim mają Reddit, Wikipedia i Google? O wszystkim dowiesz się z filmu.

3) "Atomowe nawyki" w wersji programistycznej? (film, 31 minut) https://youtu.be/WvRWQGCo5Cg INFO: Podczas prelekcji autor przedstawia cztery filary zmiany zachowania, wspomniane w książce "Atomowe nawyki", i prezentuje metodę na ich zastosowanie w pracy programisty. Celem jest wznieść swoje umiejętności programistyczne na wyższy poziom, szybciej wchodzić w stan flow i łatwiej uczyć się nowych rzeczy.

4) Uruchamianie lokalnych LLMów za pomocą Ollama i Open WebUI https://linuxtldr.com/setup-ollama-and-open-webui-on-linux/ INFO: ChatGPT jest fajny, ale wysyłanie danych firmowych do chmury OpenAI już takie fajne nie jest. Istnieje jednak dość prosta metoda na uruchomienie dowolnego (z puli open source) wybranego LLM na lokalnej maszynie i obsługiwania go przez przeglądarkę, tak jak robimy to w przypadku ChatGPT.

5) Kiedy chmura Google zawodzi - utrata danych i backupów klienta https://niebezpiecznik.pl/post/google-skasowalo-im-konto-firmowe-i-backupy-tez/ INFO: Czy można sobie wyobrazić sytuację, w której fundusz emerytalny traci dostęp do swoich danych w chmurze Google Cloud - wraz z kopiami zapasowymi? Artykuł opisuje realny przypadek australijskiego funduszu UniSuper, który musiał zmierzyć się z utratą krytycznych danych i dwutygodniową walką o odzyskanie sprawności operacyjnej. To przestroga dla wszystkich użytkowników usług chmurowych, aby nie lekceważyć konieczności posiadania zewnętrznych kopii bezpieczeństwa.

6) Nowoczesne layouty CSS - bez frameworka https://smashingmagazine.com/2024/05/modern-css-layouts-no-framework-needed/ INFO: Artykuł pokazuje, jak nowoczesne funkcje CSS pozwalają na stworzenie elastycznych i konfigurowalnych układów, bez konieczności używania jakichkolwiek frameworków.

7) Warsztaty online z architektury IT - zobacz spis! [sponsorowane] https://patoarchitekci.io/szkolenia/ INFO: Zobacz aż 7 szkoleń dla osób chcących lepiej poznać tematykę projektowania architektury aplikacji ze szczególnym uwzględnieniem platformy Azure. Pełne opisy i ceny szkoleń znajdziesz na stronie.

8) Delegowanie zadań jest łatwiejsze, gdy lepiej wyjaśniasz swoje pomysły https://newsletter.weskao.com/p/delegating-and-explaining INFO: Tytuł może wydawać się dość oczywisty, ale umiejętność efektywnego przekazywania innym swojej wizji to coś, czego można, a może i trzeba, się nauczyć. Jest to umiejętność niezbędna dla osób delegujących zadania innym. Lecz, jak robić to dobrze?

9) Ćwiczenia przeciwpożarowe vs. testy phishingowe - co je łączy? https://security.googleblog.com/2024/05/on-fire-drills-and-phishing-tests.html?m=1 INFO: Artykuł przedstawia ciekawe porównanie między historycznymi testami ewakuacyjnymi a współczesnymi testami phishingowymi, jako metodami na podnoszenie bezpieczeństwa. Zwraca uwagę na nieefektywność testów phishingowych i proponuje alternatywne podejście, które bardziej przypomina współczesne ćwiczenia przeciwpożarowe. Warto rzucić okiem.

10) Tajemnicza znikająca kropka w mailu - o SMTP słów kilka https://tjaart.substack.com/p/the-curious-case-of-the-missing-period INFO: Ciekawostka technologiczna prosto z codziennej pracy programistów: zespół twórców bada, dlaczego w treści e-maila wysłanego do jednego z klientów zniknęła kropka. Z treści drugiego maila, ten znak zniknął bez problemów. To jest interesująca historia, która ostatecznie prowadzi do nieoczywistych odkryć w specyfikacji i implementacji protokołu SMTP.

11) Lokalny zestaw narzędzi webowych do zabawy plikami PDF https://github.com/Stirling-Tools/Stirling-PDF INFO: W sieci jest mnóstwo narzędzi webowych do operowania na plikach PDF (tworzenie PDF-ów z obrazów, scalanie plików, konwersja, dzielenie na strony itp.). Ich wadą jest jednak to, że działają one 'gdzieś w chmurze', co wyklucza ich użycie w poważnych biznesach. Wrzucanie tam danych mogłoby narazić firmę na wyciek informacji. Ta aplikacja uruchamia się lokalnie i realizuje dokładnie to, co jej chmurowe odpowiedniki. Jest idealna do użytku w firmie.

12) Kompresja obrazów - jak działa? https://swistak.codes/post/kompresja-obrazow/ INFO: Artykuł opisuje techniki, które stosuje się do bezstratnej i stratnej kompresji obrazów, z dużym naciskiem na kodowanie transformacyjne DCT z JPG. Jest to dobra lektura dla osób, które lubią wiedzieć, jak coś działa.

13) PostgreSQL jako uniwersalny stack technologiczny? https://www.timescale.com/blog/how-to-collapse-your-stack-using-postgresql-for-everything/ INFO: Tworząc rozbudowaną aplikację, bardzo łatwo można popaść w piekło zależności. Kilkanaście różnych rozwiązań technologicznych obsługujących jeden moduł aplikacji. Osobna baza do użycia jako silnik wyszukiwarki, osobna jako baza do przechowywania danych, osobna do trzymania wektorów pod LLM-y itd. To tworzy nam rozbudowany stack, który kiedyś przerodzi się w dług technologiczny i tylko niepotrzebnie skomplikuje architekturę systemu. A gdyby tak wszystkie te rzeczy zastąpić Postgresem? Kilka słów o spłaszczaniu architektury aplikacji.

14) Czy to czas na drugą generacją poczty E-mail? https://gabrielsieben.tech/2024/05/17/thinking-out-loud-2nd-gen-email/ INFO: Autor dzieli się swoimi przemyśleniami na temat obecnych problemów związanych z technologią e-mail i proponuje koncepcję "drugiej generacji" poczty elektronicznej, która miałaby te problemy rozwiązywać. Materiał jest wart przemyślenia, bo nie zawsze zdajemy sobie sprawę z problemów technicznych, jakie generuje współczesna wersja e-maila.

15) Kiedy treści znikają z Internetu? https://www.pewresearch.org/data-labs/2024/05/17/when-online-content-disappears/ INFO: Czy wiesz, że 38% stron internetowych, które istniały w 2013 roku, już nie jest dostępne? Badacze przeprowadzili analizę, pokazującą niepokojący trend w znikaniu treści online. W artykule znajdziesz wyniki badań dotyczące dostępności stron internetowych z ostatniej dekady, trwałości linków na stronach rządowych i w wiadomościach, a także tego, jak często posty na mediach społecznościowych są usuwane lub stają się prywatne. Warte przemyślenia fakty na temat ulotności tego, co znajdujemy w sieci.

16) Jak wyłączyć podgląd AI w Google i ustawić 'Web' jako domyślną kartę https://tenbluelinks.org/ INFO: Google od pewnego czasu narzuca wszystkim swój dodatek do wyników wyszukiwania, którym są odpowiedzi oparte na AI. Niekoniecznie każdemu to usprawnienie przypadło do gustu. Niestety, nie istnieje oficjalna metoda na pozbycie się tej funkcji, ale dzięki specjalnemu parametrowi można wyłączyć wspomniane narzędzie. Artykuł tłumaczy, jak to osiągnąć na znanych przeglądarkach oraz urządzeniach mobilnych.

17) Klonowanie pojedynczego commitu z repozytorium GIT https://blog.hartwork.org/posts/clone-arbitrary-single-git-commit/ INFO: Sklonowanie pojedynczego commita z Gita bez jego historii nie jest wcale łatwe, a według niektórych, może nawet być niewykonalne. Autor przedstawia jednak prostą sztuczkę. Łącząc kilka mechanizmów dostępnych w Gicie, klonuje on ostatecznie jednego, samotnego commita do nowego repozytorium. To drobny trik, ale być może kiedyś przyda Ci się w pracy.

18) Doom ze zmienioną matematyką - co się stanie, gdy pi NIE będzie równe 3,14159… (film 19m) https://media.ccc.de/v/mch2022-236-non-euclidean-doom-what-happens-to-a-game-when-pi-is-not-3-14159-#t=376 INFO: Stałe trygonometryczne to coś, nad czym się nie zastanawiamy. Po prostu używamy ich w kodzie i działają. Co by się jednak stało z grą Doom, gdyby lekko zmienić wartość liczby pi w kodzie? Jak zachowałaby się gra i które mechanizmy w niej zawarte zaczęłyby szaleć? Ciekawy wykład.

19) Copilot+ PC - nowa kategoria komputerów z AI od Microsoft https://blogs.microsoft.com/blog/2024/05/20/introducing-copilot-pcs/ INFO: Microsoft wprowadza na rynek komputery osobiste z zaawansowaną sztuczną inteligencją. Te urządzenia mają zmienić sposób, w jaki na co dzień pracujemy, tworzymy i komunikujemy się, oferując niespotykane dotąd możliwości. Takie jak generowanie obrazów AI w czasie rzeczywistym czy funkcję Recall, ułatwiającą odnajdywanie informacji. Więcej informacji na stronie.

20) Narzędzia podnoszące produktywność w komunikacji z zespołami web developerów https://www.freecodecamp.org/news/how-to-collaborate-with-web-development-teams-using-productivity-tools/ INFO: Jeśli jesteś deweloperem, designerem czy PM-em w branży webowej, to z pewnością zainteresuje cię artykuł, który pokazuje, jak narzędzia do współpracy wewnątrz zespołu mogą zwiększyć produktywność całego teamu. W artykule znajdziesz omówienie kilku popularnych rozwiązań tego typu.

21) Ukrywanie vs. wyłączanie funkcji w UX - kiedy co stosować? https://smashingmagazine.com/2024/05/hidden-vs-disabled-ux/ INFO: Zastanawiasz się, kiedy lepiej ukryć, a kiedy wyłączyć czy zablokować funkcję w interfejsie użytkownika? Ten artykuł z serii o wzorcach projektowych wyjaśnia, które z rozwiązań będzie najlepsze pod względem UX dla użytkowników. Niekiedy zastosowanie nieodpowiedniego rozwiązania może doprowadzić do irytacji użytkownika.

22) Czym są arkusze SVG sprite i jak je tworzyć? https://ryantrimble.com/blog/what-the-heck-is-an-svg-sprite-sheet/ INFO: Tzw. "sprajty" to coś, co przez lata było wykorzystane w świecie grafiki webowej (jpg, png itp.) do zmniejszania liczby zapytań o elementy interfejsu i do zmniejszenia ilości zużywanego przez nie transferu. Dlaczego mielibyśmy nie zastosować tego rozwiązania do grafik wektorowych? Ten artykuł podpowie Ci, jak to zrobić.

23) Pytania do rozmowy kwalifikacyjnej z JavaScript https://www.freecodecamp.org/news/js-interview-prep-handbook/ INFO: Kolejna kompilacja pytań, często zadawanych na rozmowach technicznych związanych z JavaScriptem. Oczywiście, poza samymi pytaniami, znajdziesz tutaj także wyjaśnienia poruszanych zagadnień.

24) Wzrost emisji i zużycia wody w Microsoft z powodu AI https://www.pcmag.com/news/microsofts-emissions-spike-29-as-ai-gobbles-up-resources INFO: Rozwój sztucznej inteligencji to nie tylko postęp technologiczny, ale i wyzwanie dla środowiska. Raport Microsoftu z 2024 roku ujawnia 29% wzrost emisji i 23% większe zużycie wody, głównie przez nowe technologie AI, takie jak GPT-4.

25) Kompletny przewodnik po klasach w JavaScript https://www.freecodecamp.org/news/javascript-class-handbook/ INFO: JavaScript to nie tylko funkcje i obiekty, ale także klasy, które mogą ułatwić strukturyzację kodu i zarządzanie danymi. Ten przewodnik szczegółowo wyjaśnia działanie klas w JavaScript, omawiając, czym są pola w klasach oraz wyjaśnia zastosowanie metody 'super'.

26) Ostrożnie z użyciem ARG w Dockerfile - subtelna pułapka https://qmacro.org/blog/posts/2024/05/13/using-arg-in-a-dockerfile-beware-the-gotcha/ INFO: Ustawiasz parametry ARG w pliku Dockerfile, a one czasami działają, natomiast niekiedy zachowują się tak, jakby były ignorowane. Artykuł wyjaśnia, jak instrukcja ARG współpracuje z FROM i co zrobić, aby uniknąć problemów z pustymi wartościami ARG. Jest to praktyczna wiedza dla każdego, kto chce opanować tworzenie obrazów Dockera.

27) Mierzenie produktywności przy wytwarzaniu oprogramowania - czy to w ogóle możliwe? https://www.jamesshore.com/v2/blog/2024/a-useful-productivity-measure INFO: Czy możliwe jest wiarygodne mierzenie produktywności programistów? Zaproponowano już kilka metod na taki pomiar, ale tutaj autor przychodzi z własnym, dość innowacyjnym pomysłem. Zamiast mierzyć np. ilość commitów, linii kodu czy tasków, autor skupia się na mierzeniu odsetka czasu, który pracownik poświęca na 'zadania dające wartość'. Pracownik zaabsorbowany w 100% niekoniecznie jest produktywny w oczach firmy, a co najwyżej przeciążony zadaniami.

28) Slack pod ostrzałem za swoją politykę prywatności https://techcrunch.com/2024/05/17/slack-under-attack-over-sneaky-ai-training-policy/ INFO: Jeśli jesteś użytkownikiem Slacka, ten artykuł może Cię zainteresować. Okazuje się, że firma wykorzystuje dane użytkowników do szkolenia swoich usług AI. Jeśli nie chcesz być częścią tego procesu, musisz aktywnie z tego zrezygnować. Takie podejście do prywatności może być poważnym naruszeniem bezpieczeństwa danych firmowych, jeśli są one przekazywane w ramach tej platformy.

29) Nie 'zarządzaj w górę', a twórz zespół https://hybridhacker.email/p/dont-manage-up-team-up-instead INFO: Zamiast tradycyjnego "zarządzania z góry", które może wzmacniać hierarchię i ograniczać współpracę, autor proponuje podejście "Teaming Up". Opisuje, jak ta metoda sprzyja współdziałaniu w zespołach, jakie wyzwania niesie i jak sobie z nimi radzić. Artykuł jest przewodnikiem dla tych, którzy chcą budować bardziej przyjazne i produktywne środowisko pracy w firmie.

30) ChatGPT może rozmawiać, ale pracownicy OpenAI już nie https://www.vox.com/future-perfect/2024/5/17/24158478/openai-departures-sam-altman-employees-chatgpt-release INFO: Atmosfera wokół OpenAI gęstnieje po rezygnacji współzałożyciela i głównego naukowca. Artykuł zgłębia przyczyny ich odejścia i milczenia byłych pracowników, wiążącego się z restrykcyjnymi umowami poufności. Jak te zmiany wpłyną na przyszłość AI i transparentność działań OpenAI?

31) Udostępnij swój terminal przez internet https://github.com/tsl0922/ttyd INFO: To jest sprytne narzędzie, które umożliwia wystawienie w internecie aktualizowanego na żywo podglądu swojego terminala. Może to być użyteczne, na przykład, przy prowadzeniu wszelkiego rodzaju szkoleń.