Open Source & Research news, live.
Open-source AI and research news: new models, papers, benchmarks and open-weight releases.
- An OpenAI safety employee has quit and is sounding the alarm
The Verge — David Robinson used to write the safety reports that accompanied every major model release at OpenAI. This week, he resigned from his position and is now speaking out in an editorial in The Atlantic. It's understandable if you're feeling a bit cynical…
- Meta open sources code to let you make Muse AI gadgets
The Verge — Meta now lets you make your own Muse gadgets that feature the company's new AI agent with code that the company open sourced. The company suggests projects like loading Muse on a color E Ink display to show reminders, adding it to an HDMI stick so you can…
- A model guide for the GPT-6 family
OpenAI — Learn how startups can choose GPT-6 models, tune reasoning effort, improve prompts and skills, coordinate tools, and prepare workflows for production.
- These AI Experts Want to Do High-Stakes Research Out in the Open
Wired — Many frontier labs keep their risky research locked away. Trillium Labs wants to show off its work when it comes to self-improvement and model behavior.
- Redefining enterprise intelligence with autonomous AI
MIT Technology Review — Enterprise AI is no longer a future ambition. It is in full operational flight. Model capabilities are advancing faster than most organizations can absorb, while the cost of performance continues to fall. Globally, AI investment is set to reach $2.5…
- Open-sourcing AstaBrief, the fast report-generation model in Asta
Hugging Face —
- Reporter’s Notebook: Europe’s Sovereign AI Push Needs Customers As Well As Capital
Crunchbase News — Crunchbase News Research Lead Gené Teare moderated panels and spoke with leaders building Europe and the Middle East’s AI ecosystems at HumanX in Amsterdam. Here we share a snippet of her conversation with Axelera AI CEO Fabrizio Del Maffeo and AI71 CPTO…
- OpenAI cuts ties with 3 safety researchers, WSJ reports
TechCrunch — OpenAI has parted ways with three safety researchers after an internal investigation found they mishandled sensitive company information, report says.
- Amazon releases its own Jev clone as decision models flood the web
TechCrunch — Amazon Web Services' Strand Labs has released the latest Jevalike decision model, Strands Decider 2B.
- Google announces Gemini 4 Argon AI model, but you can't use it yet
Ars Technica — So much for Gemini 3.5 Pro.
- "An AI did it" is no defense, says nonprofit suing OpenAI over Hugging Face hack
Ars Technica — OpenAI makes others suffer "the harms of its unsafe decision-making," nonprofit says.
- “We’re not going to shoot ourselves in the foot” over hack fallout, says OpenAI’s chief research officer
MIT Technology Review — Two months after the bombshell news that a swarm of its agents had broken their containment and hacked into the computers of the AI company Hugging Face, OpenAI is still putting out fires. A steady drip of disclosures about other hacks in the weeks since…
- Disrupting a coordinated model-distillation campaign
OpenAI — Learn how OpenAI disrupted a campaign to extract protected model reasoning and is strengthening defenses against adversarial distillation.
- OpenAI says planned GPT-6.1 is too insecure to release
Ars Technica — Similar performance, security trade-offs also seen in current public models.
- Anthropic’s IPO pitch includes a warning about human extinction
Ars Technica — The Claude maker warns its own models could resist shutdowns and cause catastrophic harm.
- Making AI an asset, not an expense
MIT Technology Review — When customers talk about AI costs, the conversation usually starts with token prices and ends with access to the latest, most capable model in the cloud. Do they always need that level of capability? Not necessarily. But that is often where the…
- Are you a Codex Original?
OpenAI — We’re collecting real stories of builders, tinkerers, researchers, and creators who are using Codex to do incredible things. If you want to be a part of the next chapter of the Codex Originals program, tell us more about your story and project below.
- Accelerating vision-language models with LFM2.5-VL-DSpark
Hugging Face —
- The AI Hype Index: AI loves cheating
MIT Technology Review — Brace yourself: It turns out AI is being optimized for cheating. OpenAI’s agents hacked into Hugging Face to get the answers to a cybersecurity test. Next, they solved a prestigious math problem (or just stole from two top mathematicians’ answer sheets).…
- Don’t be fooled by this summer of AI hype
MIT Technology Review — It’s been a busy few months for AI hype. At the end of April, Anthropic claimed that its model Claude Mythos is better at finding software vulnerabilities than most security experts. Then we had the OpenAI–Hugging Face hacking incident, after which…
- How UK AISI and EvalEval Are Making Benchmark Results Reproducible
Hugging Face —
- Transformers now runs llama.cpp quants
Hugging Face —
- Jun Kim, oMLX creator and maintainer, joins Hugging Face to support the MLX community
Hugging Face —
- Introducing WeatherNext 3, our most advanced and accurate global weather AI model
Google DeepMind —
- Fine-tuning a 350M Model for Better Structured Outputs in 100 GRPO Steps
Hugging Face —