AIR#112 - June 24, 2024
Good morning, AI aficionados! As you sip your morning coffee, get ready to dive into today's edition of AIR: The AI Recon. Leading the headlines is a thought-provoking study that reveals ChatGPT's bias against resumes with disability-related credentials. This finding is stirring up important conversations about fairness and inclusivity in AI, making it a must-read for anyone interested in the ethical implications of technology. Meanwhile, researchers are making strides in improving AI accuracy by using semantic entropy to detect hallucinations in large language models, a breakthrough that promises to enhance the reliability of AI-generated responses.
But that's not all! Francois Chollet is making waves with his bold claim that large language models won't lead to artificial general intelligence (AGI) and is backing it up with a $1M prize for a true solution. This challenge has the AI community buzzing and could pave the way for groundbreaking innovations. On a lighter note, Anthropic's Claude 3.5 Sonnet is wowing AI power users with its impressive speed and capabilities, outperforming even GPT-4. This latest development is a game-changer for developers looking to push the boundaries of AI performance.
And for those who love a good ethical debate, California's new AI bill is sparking controversy among safety advocates and developers alike. While some see it as a necessary step for regulation, others fear it could stifle innovation. Whether you're here for the groundbreaking tech updates, the ethical debates, or the latest industry buzz, today's edition is packed with stories that will both intrigue and challenge you. So, sit back, sip your coffee, and let's delve into the dynamic world of artificial intelligence together!
Business
California AI Bill Sparks Controversy Among Safety Advocates and Developers
California's new AI bill sparks debate: Safety advocates see it as necessary, developers fear it will stifle innovation.
Chat with Your Wallet: InnerWallet AI Finance Tracker 💸
Chat with your wallet using InnerWallet AI for smarter finance management. Track expenses, set goals, and get insights effortlessly! 💸
Merlin Labs to Test AI-Powered KC-135 Tanker Within a Year
Merlin Labs to test AI-powered KC-135 within a year, acquiring EpiSci to enhance AI pilot capabilities for autonomous military missions.
For years many mocked Musk for believing pure vision self-driving was possible
Many mocked Musk's vision for self-driving cars, but his persistence means I no longer have to drive myself. 🚗
Gene Roddenberry Foundation Offers $1M for Ethical AI Development
Gene Roddenberry Foundation offers $1M prize for ethical AI development to benefit humanity, inspired by "Star Trek" values.
Clearview AI Settles $50M Privacy Lawsuit
Clearview AI settles privacy lawsuit for $50M, offering plaintiffs a share of future profits instead of cash payout.
Citigroup: AI Could Replace Over Half of Banking Jobs
Citigroup warns AI could replace over half of banking jobs, potentially automating 54% and augmenting 12% of roles.
How AI is Revolutionizing Warfare
AI is transforming warfare with advanced coordination, real-time data processing, and smart command systems, making forces more efficient.
Apple and Meta Discuss AI Partnership for iOS 18
Apple and Meta, longtime rivals, are discussing an AI partnership for iOS 18, potentially integrating Meta AI models into Apple Intelligence.
Amazon Eyes Monthly Fee for Alexa AI Overhaul
Amazon plans to charge $5-$10/month for a revamped Alexa with advanced AI features, aiming to turn a profit on the service.
Enveda raises $55M to combine ancient remedies with AI for drug discovery
Enveda raises $55M to harness AI and ancient remedies for drug discovery, aiming to revolutionize medicine from natural sources.
Engineering
Anthropic Employee Posts Copyrighted Claude Demo Video on YouTube
Anthropic employee posts copyrighted Claude demo video on YouTube; original creator finds work uncredited.
Anthropic's Claude 3.5 Sonnet Wows AI Power Users: 'This is Wild'
Anthropic's Claude 3.5 Sonnet outperforms GPT-4, impressing AI power users with its speed, cost, and coding abilities. "This is wild."
OpenAI's Local App to Bypass Web Permissions
OpenAI's local app bypasses web permissions, allowing AI to act on your behalf using your computer's access, revolutionizing AI automation.
[GitHub] Open-source Anthropic Artifacts UI by e2b-dev
e2b-dev releases open-source Anthropic Artifacts UI with secure AI code execution using E2B's Code Interpreter SDK.
Top Libraries to Accelerate LLM Building
Discover the top 9 libraries to streamline LLM development, from training to deployment, and boost efficiency in 2024.
[GitHub] Tsellm: Interactive SQLite Shell with LLM Integration
Integrate LLMs into your SQLite queries with tsellm! Use a simple CLI or interactive shell to enhance your database experience.
[GitHub] fchollet's Abstraction and Reasoning Corpus (ARC-AGI)
Explore fchollet's ARC-AGI on GitHub: a benchmark for testing human-like AI intelligence. 2.8k stars, 468 forks.
Claude 3.5: The First LLM for WebUI Automation [Video]
Claude 3.5 Sonnet: A game-changer in Web UI automation? Watch the image recognition test on YouTube now!
Academic
🔥 ChatGPT Biased Against Disability-Related Resume Credentials, Study Finds
Study finds ChatGPT ranks resumes with disability-related credentials lower, but bias can be reduced with customized instructions.
🔥 [Paper] Detecting Hallucinations in Large Language Models Using Semantic Entropy | Nature
Researchers use semantic entropy to detect "hallucinations" in large language models, improving accuracy by identifying unreliable answers.
Francois Chollet: LLMs Won't Lead to AGI, $1M Prize for True Solution [Video]
Francois Chollet claims LLMs won't lead to AGI, offers $1M prize for a true solution.
The Encyclopedia Project: Navigating Truth in the AI Era
Navigating AI's flood of misinformation, a family turns to encyclopedias for reliable knowledge, highlighting the struggle for truth.
AI Can't Fix Automation's Fallout
AI can't fix the harm caused by automation; it often just adds another layer of problems, hurting workers and customer service.
AI, Narcissism, and the Future of Sex
AI sex partners may reduce emotional growth, potentially increasing narcissism. The future of intimacy could be at risk.