A Daily chronicle of AI Innovations May 31 2024: ⚠️OpenAI reports misuse of its AI for ‘deceptive activity’ 🏫OpenAI introduces ChatGPT Edu for universities 💼Tech giants form industry group to develop next-gen AI chip 📱Apple planning AI-driven overhaul for Siri to manage individual apps 📚AI Jobs
A Daily chronicle of AI Innovations May 31st 2024:
⚠️OpenAI reports misuse of its AI for ‘deceptive activity’
🏫OpenAI introduces ChatGPT Edu for universities
💼Tech giants form industry group to develop next-gen AI chip
📱Apple planning AI-driven overhaul for Siri to manage individual apps
📚Perplexity launches AI publishing platform
Enjoying these updates, Listen and subscribe at AI Unraveled Podcast: AI Unraveled: Latest AI News & Trends, GPT, Gemini, Generative AI, LLMs, Prompt, AI Bedtime Stories on Apple Podcasts
⚠️OpenAI reports misuse of its AI for ‘deceptive activity’
OpenAI identified and disrupted five covert influence operations that used its generative AI models for “deceptive activity” across the internet.
OpenAI said the threat actors used its AI models to generate short comments, longer articles in various languages, made-up names, and bios for social media accounts over the last three months. These campaigns included threat actors from Russia, China, Iran, and Israel. It focused on issues including Russia’s invasion of Ukraine, the conflict in Gaza, the Indian elections, and politics in Europe and the United States, among others.
OpenAI has also published a trend analysis report that describes the behavior of these malicious actors in detail.
Why does it matter?
This provides concrete evidence of generative AI being used to spread misinformation and manipulate global affairs. While there were no novel attacks this time, defending against them in the future will require continued investment in monitoring, safeguards, and public awareness from AI developers and platforms.
Source: https://openai.com/index/openai-and-reddit-partnership
💼 Tech giants form industry group to develop next-gen AI chip
Intel, Google, Microsoft, Meta, and other tech heavyweights are establishing a new industry group called the Ultra Accelerator Link (UALink) Promoter Group to guide the development of the components that link together AI accelerator chips in data centers. The group also counts AMD, Hewlett Packard Enterprise, Broadcom, and Cisco among its members.
It is proposing a new industry standard to connect the AI accelerator chips found within a growing number of servers. It’s version one, UALink 1.0, will connect up to 1,024 AI accelerators — GPUs only — across a single computing “pod.” It is set to arrive in Q4 2024.
Why does it matter?
Nvidia, currently the biggest player in the AI chip market with an 80% share, is not in the group. Tech giants have been keen to reduce their dependence on Nvidia by working on their own custom chips. This seems to be the latest effort to break Nvidia’s dominance.
📱Apple planning AI-driven overhaul for Siri to manage individual apps
- Apple is planning a major AI upgrade for Siri in iOS 18, enabling users to control specific features within iPhone apps using voice commands, initially limited to Apple-developed apps.
- The update will eventually allow Siri to handle multiple tasks in a single request, such as summarizing a recorded meeting and sending it to a friend, as part of Apple’s larger AI strategy to be unveiled at WWDC on June 10.
- New AI-driven capabilities will enable Siri to control iPhones more precisely, including opening files, moving notes, managing emails, and summarizing articles, enhancing device navigation and functionality.
- Source: https://www.newsbytesapp.com/news/science/siri-to-gain-enhanced-app-control-in-ios-18/story
Perplexity launches AI publishing platform
- Perplexity introduced ‘Pages’, an AI-driven feature that creates customizable webpages for research and writing based on user prompts.
- The ‘Pages’ tool gathers information using Perplexity’s AI models, organizes it into sections with citations, and tailors content for different audience levels but requires new prompts to correct errors.
- The Verge tested ‘Pages’ on a complex topic and noted its difficulty with intricate subjects, showing it might struggle with in-depth research despite effectively explaining fundamental concepts.
- Source: https://www.newsbytesapp.com/news/science/perplexity-launches-new-ai-feature-called-pages/story
🏫OpenAI launches ChatGPT Edu for universities
- OpenAI has introduced ChatGPT Edu, a new version of ChatGPT, aimed at improving the academic and operational efficiency of universities with advanced text interpretation and data analysis capabilities, and support for over 50 languages.
- This version was created following the success of ChatGPT Enterprise in academic institutions such as Oxford, Wharton, and ASU, where it has been used for tasks ranging from large dataset analysis to personalized language practice.
- ChatGPT Edu provides affordable access to the GPT-4o model, enhanced message limits, robust security measures, and assures that data and conversations will not be used for training OpenAI models, ensuring a safe environment for educational use.
- Source: https://www.maginative.com/article/openai-launches-chatgpt-edu-for-universities/
OpenAI says it stopped multiple covert influence operations that abused its AI models. https://www.engadget.com/openai-says-it-stopped-multiple-covert-influence-operations-that-abused-its-ai-models-225115466.html
Google says it fixed the AI Overviews everyone has been roasting for weeks. https://bgr.com/tech/google-says-it-fixed-ai-overviews-blames-some-of-the-problems-on-you/
Anthropic’s Claude AI now autonomously interacts with external data and tools. https://venturebeat.com/ai/anthropic-claude-ai-now-autonomously-interacts-with-external-data-and-tools/
🎬The Simulation unveils ‘Netflix of AI’
AI entertainment startup The Simulation (formerly Fable Studio) just launched Showrunner, a platform that allows users to generate and watch AI-powered TV shows set in virtual simulated worlds.
- Showrunner combines multi-agent simulations with LLMs to create interactive content aligned with specific virtual worlds.
- Users can watch, direct, and star in shows set in virtual environments populated by AI characters, with the ability to craft episodes from prompts.
- The platform is launching with 10 original shows, with users able to generate new episodes and edit deeper into scripts, shots, and voices.
- Showrunner is opening in alpha to a limited number of users, with select user-created episodes receiving payment, rev share, and IMDB credits.
- The studio went viral last year after the release of its South Park episode demos and research paper on its SHOW-1 model.
Showrunner is a wild step towards the merging of AI, gaming, and traditional entertainment, enabling users to not only consume but create stories within simulated worlds. The lines are blurring fast between creators and audiences — and the traditional Hollywood media model may never be the same.
Source: https://x.com/fablesimulation/status/1796245808066740274
New AI Job Opportunities on May 31st 2024 Waymo — Software Engineer, Computer Vision/Deep Learning: https://jobs.therundown.ai/jobs/60918768-software-engineer-computer-vision-deep-learning Notable — Strategic Account Executive Findem — Director of Sales, Enterprise Twelve Labs — Lead Software Engineer, Frontend
What Else Is Happening in AI on May 31st 2024
Anthropic’s Claude can now autonomously interact with external data and tools
Anthropic has announced the general availability of Tool Use for its AI, Claude. It allows Claude to autonomously interact with external data sources, APIs, and tools, making it a business-ready trusted AI solution that can be confidently adopted by enterprises across various industries. (https://venturebeat.com/ai/anthropic-claude-ai-now-autonomously-interacts-with-external-data-and-tools)
Perplexity goes beyond AI search, launches publishing platform ‘Pages’
It is a tool to transform disorganized AI knowledge and research into easily digestible articles and reports for sharing. Users just have to describe the topic, select the target audience, and it will produce an in-depth article with a clear title, subheadings, associated media, and relevant citations. (https://x.com/perplexity_ai/status/1796203494401040846)
Google poaches AWS and Microsoft executives to strengthen AI offerings for cloud customers
Google hired two executives from these rivals as it builds out its AI ranks with the goal of boosting its cloud business. Saurabh Tiwary, a former corporate VP at Mircosoft, will join a newly created role as general manager of cloud AI. Raj Pai, a former VP at AWS, will oversee product management of the cloud AI team. (https://www.theinformation.com/articles/google-poaches-aws-microsoft-executives-for-cloud-unit)
Scale AI introduces SEAL Leaderboards, a new evaluation metric for frontier AI models
Trusted third-party evaluations are a missing part of the AI ecosystem, which is why Scale AI built them. These evals will be impossible to overfit, private, domain expert evaluations, unexploitable, and continuously updated with new data and models. (https://x.com/alexandr_wang/status/1795857651592491281)
OpenAI offers nonprofit organizations discounts on corporate ChatGPT subscriptions
It seeks to grow sales of its AI product to enterprises. Under the program, large nonprofits can get 50% off the enterprise-grade version of ChatGPT. Smaller nonprofits using ChatGPT Team will pay $20 per month per user instead of $25 or $30. (https://www.reuters.com/technology/openai-offers-nonprofits-discounts-corporate-chatgpt-product-2024-05-30)
