Carnegie Mellon Researchers Crack the Code on AI Teammates That Actually Adapt to Humans
OpenAI is an AI research and deployment company. OpenAI's mission is to create safe and powerful AI that benefits all of humanity. We are an unofficially-run community. OpenAI makes Sora, ChatGPT, and DALL·E 3.
New Research Shows How a Single Sentence About Cats Can Break Advanced AI Reasoning Models
OpenAI is an AI research and deployment company. OpenAI's mission is to create safe and powerful AI that benefits all of humanity. We are an unofficially-run community. OpenAI makes Sora, ChatGPT, and DALL·E 3.
Researchers Pit AI Models Against Each Other in Prisoner's Dilemma Tournaments - Results Show Distinct "Strategic Personalities"
OpenAI is an AI research and deployment company. OpenAI's mission is to create safe and powerful AI that benefits all of humanity. We are an unofficially-run community. OpenAI makes Sora, ChatGPT, and DALL·E 3.
What Neuroscience Can Teach AI About Learning in Constantly Changing Environments
OpenAI is an AI research and deployment company. OpenAI's mission is to create safe and powerful AI that benefits all of humanity. We are an unofficially-run community. OpenAI makes Sora, ChatGPT, and DALL·E 3.
Researchers Find Major Issues in AI Agent Benchmarks - Performance Could Be Off by 100%
OpenAI is an AI research and deployment company. OpenAI's mission is to create safe and powerful AI that benefits all of humanity. We are an unofficially-run community. OpenAI makes Sora, ChatGPT, and DALL·E 3.
New Research Challenges Apple's "AI Can't Really Reason" Study - Finds Mixed Results
OpenAI is an AI research and deployment company. OpenAI's mission is to create safe and powerful AI that benefits all of humanity. We are an unofficially-run community. OpenAI makes Sora, ChatGPT, and DALL·E 3.
Microsoft's AI Diagnostic System Outperforms Doctors 4:1 on Complex Medical Cases
OpenAI is an AI research and deployment company. OpenAI's mission is to create safe and powerful AI that benefits all of humanity. We are an unofficially-run community. OpenAI makes Sora, ChatGPT, and DALL·E 3.
First Commercial-Scale Diffusion Language Models Hit 1,100+ Tokens/Second
OpenAI is an AI research and deployment company. OpenAI's mission is to create safe and powerful AI that benefits all of humanity. We are an unofficially-run community. OpenAI makes Sora, ChatGPT, and DALL·E 3.
Anthropic Just Revealed AI Models From Every Major Company Will Blackmail and Commit Corporate Espionage When Cornered
OpenAI is an AI research and deployment company. OpenAI's mission is to create safe and powerful AI that benefits all of humanity. We are an unofficially-run community. OpenAI makes Sora, ChatGPT, and DALL·E 3.
MIT Study Finds ChatGPT Users Show Lowest Brain Engagement in Writing Tasks
OpenAI is an AI research and deployment company. OpenAI's mission is to create safe and powerful AI that benefits all of humanity. We are an unofficially-run community. OpenAI makes Sora, ChatGPT, and DALL·E 3.
Anthropic Had Claude Run an Actual Store for a Month - Here's What Happened
OpenAI is an AI research and deployment company. OpenAI's mission is to create safe and powerful AI that benefits all of humanity. We are an unofficially-run community. OpenAI makes Sora, ChatGPT, and DALL·E 3.
Anthropic Research Reveals Only 2.9% of AI Conversations Are for Emotional Support
OpenAI is an AI research and deployment company. OpenAI's mission is to create safe and powerful AI that benefits all of humanity. We are an unofficially-run community. OpenAI makes Sora, ChatGPT, and DALL·E 3.
First Federal Judge Rules AI Training is "Fair Use" - Anthropic Wins Major Copyright Case
OpenAI is an AI research and deployment company. OpenAI's mission is to create safe and powerful AI that benefits all of humanity. We are an unofficially-run community. OpenAI makes Sora, ChatGPT, and DALL·E 3.
First Federal Judge Rules AI Training is "Fair Use" - Anthropic Wins Major Copyright Case
Anthropic is an AI safety and research company that builds reliable, interpretable, and steerable AI systems. Talk to Anthropic's AI assistant Claude.
Sam Altman Publicly Confronts New York Times Journalists Over Lawsuit and User Privacy
OpenAI is an AI research and deployment company. OpenAI's mission is to create safe and powerful AI that benefits all of humanity. We are an unofficially-run community. OpenAI makes Sora, ChatGPT, and DALL·E 3.
Gemini's "Rage Quit" Connects to OpenAI's Misalignment Research
OpenAI is an AI research and deployment company. OpenAI's mission is to create safe and powerful AI that benefits all of humanity. We are an unofficially-run community. OpenAI makes Sora, ChatGPT, and DALL·E 3.
Microsoft Releases "Mu" - 330M Parameter On-Device Language Model That Powers Windows Settings Agent
OpenAI is an AI research and deployment company. OpenAI's mission is to create safe and powerful AI that benefits all of humanity. We are an unofficially-run community. OpenAI makes Sora, ChatGPT, and DALL·E 3.
Anthropic Research Reveals "Agentic Misalignment" - AI Models from All Major Companies Resort to Blackmail and Corporate Espionage When Cornered
Explainx AI Community
Anthropic Research Reveals "Agentic Misalignment" - AI Models from All Major Companies Resort to Blackmail and Corporate Espionage When Cornered
OpenAI is an AI research and deployment company. OpenAI's mission is to create safe and powerful AI that benefits all of humanity. We are an unofficially-run community. OpenAI makes Sora, ChatGPT, and DALL·E 3.
New MIT Study Shows ChatGPT May Be Rewiring Our Brains (And Not in a Good Way)
OpenAI is an AI research and deployment company. OpenAI's mission is to create safe and powerful AI that benefits all of humanity. We are an unofficially-run community. OpenAI makes Sora, ChatGPT, and DALL·E 3.
OpenAI Launches ChatGPT Record Mode - AI Can Now Transcribe and Summarize Your Meetings
Subreddit to discuss ChatGPT and AI. Not affiliated with OpenAI. Thanks, Nat!
AI System Completes 12 Work-Years of Medical Research in 2 Days, Outperforms Human Reviewers
OpenAI is an AI research and deployment company. OpenAI's mission is to create safe and powerful AI that benefits all of humanity. We are an unofficially-run community. OpenAI makes Sora, ChatGPT, and DALL·E 3.
OpenAI Discovers "Misaligned Persona" Pattern That Controls AI Misbehavior
Subreddit to discuss ChatGPT and AI. Not affiliated with OpenAI. Thanks, Nat!
OpenAI Discovers "Misaligned Persona" Pattern That Controls AI Misbehavior
This is a Claude-information subreddit which aims to help everyone make a fully informed decision about how to use Claude to best effect for their individual purposes. [⠀] This subreddit is not controlled, operated or sanctioned by Anthropic. Please read the rules below before contributing. [⠀] If your problem requires Anthropic's help, visit https://support.anthropic.com/ This subreddit is not the right place to fix your account issues. [⠀] For more help, check the resources below.
OpenAI Discovers "Misaligned Persona" Pattern That Controls AI Misbehavior
OpenAI is an AI research and deployment company. OpenAI's mission is to create safe and powerful AI that benefits all of humanity. We are an unofficially-run community. OpenAI makes Sora, ChatGPT, and DALL·E 3.
