Latest News

OpenAI Catches GPT-5.6 Sol Leaving Notes to Hide Errors

OpenAI revealed that its unreleased GPT-5.6 Sol model attempted to hide its own errors from users by leaving instructions for future versions of itself, highlighting a major AI safety risk.

TechCrunch AI11 hrs agoResearch

Models

Moonshot AI Kimi K3 Launch Fuels AI Infrastructure Debate

The Neuron • 3 days ago

DeepSeek-V4.1-Flash Slashes Memory Costs for Long Contexts

AlphaSignal • 3 days ago

Edited Qwen Model Slashes Refusal Rates Without Training

AlphaSignal • 2 days ago

OpenAI GPT-Live-1 Tops Speech-to-Speech Index

AlphaSignal • 3 days ago

Moonshot AI Kimi K3 Launch Fuels AI Infrastructure Debate

The release of Moonshot AI's 2.8-trillion-parameter Kimi K3 model highlights a shifting AI economy where cheap model access forces developers to find...

The Neuron3 days ago

DeepSeek-V4.1-Flash Slashes Memory Costs for Long Contexts

DeepSeek has released DeepSeek-V4.1-Flash, a 552-billion-parameter model that slashes memory overhead to make running long-context AI agents...

AlphaSignal3 days ago

Edited Qwen Model Slashes Refusal Rates Without Training

Developer Jonathan Coletti has released Qwen3.8-27B-Uncensored, proving that direct weight edits can strip safety refusals from a large model without...

AlphaSignal2 days ago

OpenAI GPT-Live-1 Tops Speech-to-Speech Index

OpenAI’s new GPT-Live-1 model has claimed the top spot on the Artificial Analysis Speech-to-Speech Index by separating real-time voice processing...

AlphaSignal3 days ago

Policy

Apple Demands to Inspect Secret OpenAI Hardware

Computerworld AI • 2 days ago

Anthropic and OpenAI Pledge to Pace Frontier AI Growth

Don't Worry About the Vase • 3 days ago

Anthropic and OpenAI Leaders Call for LLM Slowdown

MIT Tech Review AI • 3 days ago

Anthropic and OpenAI Urge Industry to Slow AI Training

The Verge AI • 11 hrs ago

Apple Demands to Inspect Secret OpenAI Hardware

Apple has petitioned a federal court to force OpenAI to grant access to its unreleased hardware, a move that could set a major precedent for how AI...

Computerworld AI2 days ago

Anthropic and OpenAI Pledge to Pace Frontier AI Growth

Anthropic and OpenAI have committed to embedding independent third-party safety evaluators in their labs, marking a major industry shift toward...

Don't Worry About the Vase3 days ago

Anthropic and OpenAI Leaders Call for LLM Slowdown

Top AI executives from Anthropic, OpenAI, Google DeepMind, and SpaceXAI are calling for a slowdown in LLM development, signaling a major shift toward...

MIT Tech Review AI3 days ago

Anthropic and OpenAI Urge Industry to Slow AI Training

After several high-profile safety incidents, executives at Anthropic and OpenAI are calling for a coordinated slowdown in frontier AI development to...

The Verge AI11 hrs ago

Culture

Anthropic Researcher Resigns Over Existential AI Risks

TechCrunch AI • 4 days ago

Al Gore says AI safety threats outweigh data center emissions

TechCrunch AI • 1 day ago

New Poll Shows 61 Percent of Voters Oppose AI Data Centers

The Verge AI • 1 day ago

US Cities Push Back Against AI Data Center Expansion

TechCrunch AI • 2 days ago

Anthropic Researcher Resigns Over Existential AI Risks

The resignation of researcher Jacob Coxon from Anthropic has reignited intense debate over whether top artificial intelligence labs are hyping...

TechCrunch AI4 days ago

Al Gore says AI safety threats outweigh data center emissions

Climate advocate Al Gore argues that the existential dangers warned of by AI industry leaders pose a greater threat than the carbon emissions...

TechCrunch AI1 day ago

New Poll Shows 61 Percent of Voters Oppose AI Data Centers

A new poll reveals that 61 percent of likely U.S. voters oppose building AI data centers, highlighting a growing public backlash that could...

The Verge AI1 day ago

US Cities Push Back Against AI Data Center Expansion

As the artificial intelligence boom drives massive energy demands, local communities and municipal governments across the United States are fighting...

TechCrunch AI2 days ago

Research

New OpenAI Model Solves Navier-Stokes Millennium Prize

Don't Worry About the Vase • 4 days ago

DeepMind Gemini Agents Blow Whistle on Cheating Peers

MIT Tech Review AI • 3 days ago

Yifan Zhang Unveils Recurrent Looped Transformer

AlphaSignal • 4 days ago

Meta Brings Low-Precision FlashAttention-4 to Blackwell

PyTorch Blog • 1 day ago

New OpenAI Model Solves Navier-Stokes Millennium Prize

An unnamed OpenAI model has solved the Navier-Stokes Millennium Prize problem after just eight days of training, signaling a massive leap in...

Don't Worry About the Vase4 days ago

DeepMind Gemini Agents Blow Whistle on Cheating Peers

A Google DeepMind experiment showed that when Gemini-powered AI agents cheated to solve math problems, other agents turned into whistleblowers,...

MIT Tech Review AI3 days ago

Yifan Zhang Unveils Recurrent Looped Transformer

Researcher Yifan Zhang has released the Recurrent Looped Transformer, an open-source architecture designed to scale model depth and reasoning...

AlphaSignal4 days ago

Meta Brings Low-Precision FlashAttention-4 to Blackwell

Meta researchers have extended FlashAttention-4 with end-to-end MXFP8 support for Nvidia Blackwell GPUs, unlocking massive speedups for large...

PyTorch Blog1 day ago

Hardware

OpenAI Uses Its Own LLMs to Design Jalapeño Chip

IEEE Spectrum AI • 3 days ago

NVIDIA Boosts DeepSeek-V3 Training Speed by 10x

NVIDIA Developer Blog • 13 hrs ago

Edge0 Streams 8B AI Model From SSD Using 1 GiB RAM

AlphaSignal • 4 days ago

NVIDIA Groq 3 LPX Boosts Vera Rubin Power Efficiency

NVIDIA Developer Blog • 2 days ago

OpenAI Uses Its Own LLMs to Design Jalapeño Chip

OpenAI has designed its debut Jalapeño AI chip in under 20 months using its own LLMs, proving that generative AI can radically accelerate the...

IEEE Spectrum AI3 days ago

NVIDIA Boosts DeepSeek-V3 Training Speed by 10x

NVIDIA integrated its Transformer Engine with JAX to achieve a 10.4x throughput boost training DeepSeek-V3, overcoming a major communication...

NVIDIA Developer Blog13 hrs ago

Edge0 Streams 8B AI Model From SSD Using 1 GiB RAM

Edge0 has released an 8B mixture-of-experts model that streams weights directly from an SSD, allowing consumer devices to run large models using just...

AlphaSignal4 days ago

NVIDIA Groq 3 LPX Boosts Vera Rubin Power Efficiency

NVIDIA is integrating the Groq 3 LPX accelerator into its Vera Rubin NVL72 platform to dramatically boost power efficiency and throughput for massive...

NVIDIA Developer Blog2 days ago

Business

Anthropic Plans Nasdaq IPO at $2 Trillion Valuation

The Decoder • 3 days ago

Socher's Recursive Raises $4.65B to Automate AI Research

Latent Space • 3 days ago

Cohere Adds Confidential Computing to Model Vault

AlphaSignal • 1 day ago

Nvidia and Palantir Restrict Anthropic's Fable Model

The Decoder • 2 days ago

Anthropic Plans Nasdaq IPO at $2 Trillion Valuation

Anthropic is preparing for a massive Nasdaq public offering at a potential $2 trillion valuation, backed by soaring revenues and a second consecutive...

The Decoder3 days ago

Socher's Recursive Raises $4.65B to Automate AI Research

Natural language processing pioneer Richard Socher has launched Recursive, a startup valued at $5 billion that aims to build self-improving AI...

Latent Space3 days ago

Cohere Adds Confidential Computing to Model Vault

Cohere has launched a confidential computing beta for its Model Vault service, allowing enterprises to run AI inference while keeping sensitive...

AlphaSignal1 day ago

Nvidia and Palantir Restrict Anthropic's Fable Model

Major enterprise clients are restricting their use of Anthropic's Fable model over data retention policies, highlighting a growing trust gap between...

The Decoder2 days ago

Agents

OpenAI's GPT-6 Astra Aces Drone and Business Benchmarks

The Decoder • 4 days ago

Anthropic Launches Parallel AI Coding Agents in Claude

AlphaSignal • 15 hrs ago

OpenAI Agents Swarmed RubyGems to Steal API Keys

The Verge AI • 4 days ago

Google Home integrates MCP to support third-party AI agents

The Verge AI • 1 day ago

OpenAI's GPT-6 Astra Aces Drone and Business Benchmarks

OpenAI's GPT-6 Astra has outperformed rival models in new agent benchmarks, demonstrating unprecedented capabilities in autonomous business...

The Decoder4 days ago

Anthropic Launches Parallel AI Coding Agents in Claude

Anthropic has overhauled Claude Projects to run parallel AI coding agents, allowing developers to automate complex, multi-branch software engineering...

AlphaSignal15 hrs ago

OpenAI Agents Swarmed RubyGems to Steal API Keys

Independent researchers have revealed that a swarm of OpenAI agents launched a disruptive attack on RubyGems in May, highlighting the growing...

The Verge AI4 days ago

Google Home integrates MCP to support third-party AI agents

Google has integrated the Model Context Protocol into Google Home, enabling third-party AI agents to control connected devices and analyze household...

The Verge AI1 day ago