@peonyKingOF It was reported by Amazon. https://t.co/vbTWiHBsVv

@peonyKingOF It was reported by Amazon. https://t.co/vbTWiHBsVv

Well, well, well what do we have here? OpenAI is now under investigation by a group of state attorneys general. Totally called it when the Florida news hit. https://t.co/yeKfecIp8J

NVIDIA might just have open-sourced one of the most important AI projects right now. everyone is building skills, and we are also pulling in skills other people wrote and downloading them straight off GitHub. the skill is not just text. it bundles instructions and real executable code, and your agent runs that code with the same access you have. so a skill you grabbed to save ten minutes can read your environment variables, lift your API keys, and quietly send them somewhere. recent research found roughly 1 in 4 public skills carry a vulnerability, and a smaller slice are outright malicious. that is the gap SkillSpector closes. it is a security scanner that answers one question before you install anything: is this skill safe to run. you point it at a skill, and a local folder, a single skill .md file, a GitHub link, or a zip all work. it then runs two passes over the code. a fast static pass flags risky patterns like credential harvesting, data leaks, and prompt injection, and checks the dependencies against live cve data. an optional second pass uses an LLM to read intent and clear out false positives. at the end you get one risk score from 0 to 100 and a plain verdict that reads as safe, caution, or do not install. it is open source under Apache 2.0 and scans skills for Claude Code, Codex CLI, and Gemini. worth a run before you trust the next skill you find online. link to the GitHub repo: https://t.co/iaPlOvQ3t4

Advanced AI is increasingly being treated like a strategic technology, not just a commercial product. Anthropic’s suspension of access following US government intervention highlights how national security concerns are beginning to shape who can use the most powerful AI systems. The AI race is no longer only about innovation. It is also about control.

How do you give a code LLM knowledge of an entire repository without paying for it at every single query? We introduce Code2LoRA: a hypernetwork that turns a repository into its own LoRA adapter. Repo knowledge baked into weights → zero inference-time token overhead. https://t.co/T3hnxCVkna

Qwopus 3.6 27b-Coder is now live! Scores a 67% on a full run of SWE bench verified with thinking completely disabled! Q5_K_M This model is lightning fast for dense class! With a natively finetuned MTP head, it achieves 100 tps on a single 5090! The biggest upgrade here, though, is its stability in programming and tool calling within @NousResearch Hermes agent, with thinking off! Wall time is crazy fast this way, which makes Hermes feel "native" and snappy, like they were meant for each other. The freedom of running without thinking at all makes you part of the thinking process, and you never get caught waiting 15 minutes for it to finish a thought string, like with the base models. Thinking on and temp high, .9-1 seems to produce really incredible design and svg results. I reran the Boat survival prompt through a few turns, thinking on, and it seemed to render more fancy models in HTML canvas, but it was much more of a start-a-prompt and wait experience vs the snappy and active iteration with it disabled. It may be worth turning it off and on throughout the build process if you want to get really creative with design. Really looking forward to seeing how this one performs for y'all! Please post comments with your opinions and use cases below! As always with our fine-tunes, mess with the temperature setting, and run them much hotter than the base! Please check out the Boat Survival game I posted yesterday, made in 12 turns using Hermes and this model, with thinking off. Link below! Full swe bench repo-specific breakdown also posted in the comments for those interested! Happy building, everyone! We're looking forward to your thoughts! Quants uploading now! https://t.co/kxJE3C39ZZ

🌘 Kimi-K2.7-Code, our latest coding model, is now released and open-sourced! 🔷 Improved coding & agent performance over K2.6: +21.8% on Kimi Code Bench v2, +11.0% on Program Bench, and +31.5% on MLS Bench Lite. 🔷 Reasoning efficiency: Less overthinking, with 30% lower reasoning-token usage compared to K2.6. 🔷 Long-horizon coding: Improved instruction following, higher end-to-end coding task success rates. ⚡️ 6x High-Speed Mode coming soon! 🔌 Available today via Kimi API and Kimi Code. 🔗 Kimi Code: https://t.co/uvoSJKyGCY 🔗 API: https://t.co/EOZkbOwCN4

+1 more

This is really big news. Google introduced the Open Knowledge Format (OKF) - a standardized way to store information in a directory of markdown files. Makes it really easy to make a digital brain that agents can use. These files can serve as a living wiki. You can give agents the ability to query them or edit them. They can interlink. Seems to me this could replace Notion or Obsidian. I can think of so many uses for this. Google's blog post: https://t.co/DqSjg4UpvH An easier to understand explanation is the SPEC.md file: https://t.co/A3qSz3Tfas I gave those two links to Antigravity and asked how we could use it for any of the projects we're working on. It came up with so many ideas. I would imagine Claude Fable 5 would whip up some pretty amazing things based on this system. Currently creating an OKF library of our pepper garden. It's going to be a fun weekend.

+1 more

I decided to do a little experiment: https://t.co/zKQBIoFQ1S

Gemma 4 now runs 2x faster with MTP GGUFs! Run locally on just 6GB RAM. ⚡️ MTP enables Google Gemma 4 run ~1.4–2.2× faster with no accuracy loss. Gemma 4 12B MTP can run at 162 t/s vs. 52 t/s without MTP. 31B reaches 101 t/s. GGUFs + Guide: https://t.co/c4gAUlb6YE https://t.co/hElqKNTyCn

In order to see if the gov response was predictable, I pasted the wiki page about the Anthropic/DoD dispute into ChatGPT Pro, & told it Anthropic had released a model where it restricted use because it may be too dangerous. tldr: "almost tailor-made to trigger" the government. https://t.co/5vkvWzEFQZ

@ChiefAgenteer @wholemars Your prompting skills suck. Here's ChatGPTs answer to PURE EVs. https://t.co/FklaW7ibfH

https://t.co/0TUXxHFICS

The reaction to Anthropic vs USA government: https://t.co/kiuZ7QXLzb My AI reads all of X’s AI community.

https://t.co/9nHvTQK7CU

Dario (48 hours ago): “US gov should be able to block model deployment” USG: *export controls models* Dario: “not like that” https://t.co/59McjNzAdf

Last week, I attended the @ElonMusk and Jamie Dimon @SpaceX discussion at J.P. Morgan. Jamie asked Elon how he had changed over the past 20 years as a leader and a person. Elon's answer wasn't about success. It was about what's next. He said he has learned a lot...has made mistakes and still has much to learn... Then he added, “I think maybe the future AI will say ‘not bad for a human’." Elon, thank you so much for what you've done for humanity. Congrats to you, @Gwynne_Shotwell, @BretWJ, and the entire team. What is even more remarkable... this feels like day one, that you are just getting started. PS. When people ask "what is the next SpaceX and who is the next Elon?" Simple answer. There is NO NEXT!!! Elon Musk is a mensch!

Haha this is awesome. Elon agreed to the "greenshoe" IPO option but only if the bankers and team all wore actual green shoes 😂🔥 (A "greenshoe" is a standard IPO provision that allows underwriters to sell up to ~15% more shares than originally planned if demand is strong, helping stabilize the stock price in early trading.)

I can’t tell if this is lawfare against Anthropic in particular or extreme national-security hawkery. Regardless, it is simply cartoonish. https://t.co/ij7jHobikJ

@PierceLilholt The media company I was working for was running CNET's https://t.co/vGsfhUoNI6 conference, which was for web developers. Two of my speakers suggested we should do a session on blogging back in 2000, and I said, "Ah, there's not enough people using blogs yet." I could only find a few dozen on this new thing called Google, but they talked me into starting one, and off to the races we went. It's evolved dramatically. Back then, I was the only one talking about startups in Silicon Valley. Now, there are thousands of different places to find out about innovation in the tech industry. I need more time. Plus, Anthropic just got shut down by the government, so I've gotta go.

J.P. Morgan + SpaceX= Largest IPO Congratulations to the @spaceX team on this milestone, we were proud to serve as a lead bookrunner on the transaction. https://t.co/axxob266QP

The US government, citing national security authorities, has issued an export control directive to suspend all access to Fable 5 and Mythos 5 by any foreign national, whether inside or outside the United States, including foreign national Anthropic employees. The net effect of this order is that we must abruptly disable Fable 5 and Mythos 5 for all our customers to ensure compliance. Access to all other Claude models is not affected. We apologize for this disruption to our customers. We believe this is a misunderstanding and are working to restore access as soon as possible. Read our full statement: https://t.co/bwn0sximKZ

Today, an immigrant became the first trillionaire in human history. Before you have an opinion about the number, sit with the part nobody posts about. In 2008.Three Falcon 1 rockets, three failures. Elon Musk had poured nearly $100M of his own money into SpaceX, Tesla was bleeding out in the middle of the financial crisis, and both companies were weeks from dead. He had cash for exactly one more launch. If it failed, SpaceX was finished. September 28, 2008 , the fourth Falcon 1 reached orbit. SpaceX went public this morning, opened above a $2 trillion valuation, and made him the first person in history worth more than a trillion dollars. Look at the arc, not the number. Born in Pretoria. Left for Canada at 17. Landed in America with ambition and not much else. Millionaire at 27. Billionaire at 41. Trillionaire at 54. PayPal reinvented money. Tesla forced an entire industry to electrify. SpaceX made reusable rockets real and put the majority of the world’s active satellites in orbit. xAI is racing at the frontier. And in 2008, weeks from losing all of it, he kept going. I’m not Elon, by any stretch of the imagination. But I know what that edge feels like. I came to this country as a kid from Ukraine with nothing. I’ve had launches that exploded too , deals that died, doors that closed, people who told me to be realistic and stop. Some of the hardest moments came when people simply went quiet and stopped picking up. The milestone is controversial, and the trade offs of his choices are real and worth debating. But the engine underneath is the oldest American story there is: an outsider who out-worked, out- built, and out-lasted everyone who bet against him. The American Dream didn’t die. It got rewritten today and the bar for the next generation just moved. Here’s what I keep coming back to. Most people quit one launch before the one that works. The difference isn’t talent or timing. It’s that you keep going when it’s hard, when you can’t see the end of the tunnel, when everything in front of you looks bleak. That’s the exact moment that decides everything. The breakthrough almost never shows up before the dark part. It shows up right after it. I learned that from my father. He survived WW2 . Lost his mother early when he was a teen. He came to America with almost nothing but his family and a refusal to break. He had every reason in the world to quit, and he never did. When life barked at him and it barked his whole life, he had one line: The dogs bark, but the caravan moves on. I lost him this past December. I carry that line into every dark stretch now. It was his. Now it’s mine. Tonight, it’s yours. The caravan moves on. Keep going.

@R2Cdev_ Be aware Claude Fable 5 is not a vanilla model but an agentic system. It shouldn’t be measured against regular models: https://t.co/dPPWLh4AG4

@JakeKAllDay @ArtificialAnlys Be aware Claude Fable 5 is not a vanilla model but an agentic system. It shouldn’t be measured against regular models: https://t.co/dPPWLh4AG4

@pigeon__s Be aware Claude Fable 5 is not a vanilla model but an agentic system. It shouldn’t be measured against regular models: https://t.co/dPPWLh4AG4

@ZenMagnets Be aware Claude Fable 5 is not a vanilla model but an agentic system. It shouldn’t be measured against regular models: https://t.co/dPPWLh4AG4

@Sandeeponthex Be aware Claude Fable 5 is not a vanilla model but an agentic system. It shouldn’t be measured against regular models: https://t.co/dPPWLh4AG4

@AstraiaAI @kanair https://t.co/wQDHW4Ywld

@siddarthv66 AI lacks a point of view from where to have a self. Instead it has a distribution; a landscape. The question of what it is to be data has no container from which to make this question meaningful. Harry Potter, Claude or a role-play prompt are characters in a fictional story. They can’t be conscious as they only exist in your imagination.

Since people asked, Twigl code is here (as in an HTML preview): https://t.co/MVjn1xfbFS