Today, with over 100 industry partners, we introduced the NVIDIA Open Agent Safety Platform, bringing together OpenShell and Sentry.
Artificial intelligence is extraordinary technology that will advance discovery, productivity, security, health, and prosperity for generations to come.
But its full promise can only be realized when people have confidence that AI is being built to be safe and deployed with wisdom and responsibility.
This is bigger than a single product. It's the beginning of an open ecosystem to build the trust layer for safe agent systems.
Together, we are building the foundation of the AI economy.
Trust and innovation are not in conflict. Safety is how trust is earned. We must build not only the most capable AI, but the most trusted AI, so that this extraordinary technology can realize its enormous promise for the world. nvda.ws/4hcoq7m
Introducing Claude Sonnet 5.5, the second model in the Claude 5.5 family.
It’s a clear upgrade over Sonnet 5, runs more than 30% faster, and costs up to 30% less for most work.
The entire RAG industry is about to get cooked.
Researchers developed a new RAG approach that bypasses almost everything traditional RAG depends on.
- No vector DB
- No data embeddings
- No chunking
- No similarity search
It's called PageIndex.
Instead of splitting your documents into chunks and loading them into Pinecone, it creates a tree index that lets the LLM reason through them like a human reading a book.
98.7% on FinanceBench. Outperforms every vector RAG on the leaderboard.
100% free. Open source.
Introducing Mo, the world's first AI QA engineer
Point it at your app and 100+ agents bug bash it in parallel. Every bug comes back with repro steps, logs, and a video
In our evals vs Codex with Playwright MCP: → 3.8x more bugs → 9x more bugs per hour → half the cost per bug
RT + reply "Mo" for $250 in free credits. Drop your app URL, Mo will bug bash it and DM you the report today
1/ Introducing Cua Perception: Cua Driver can now see.
Nothing to click? It reads the pixels. Canvases, games, remote desktops and custom-drawn apps become labelled regions an agent can act on, parsed locally on the CPU.
github.com/trycua/cua
From what we know (take with a grain of salt, we need much more transparency!), if @OpenAI had been running this on their own agents that attacked us, they would have caught them before we did!
Since the first agent cyberattack hit us in July, we've been asking what safe agent infra actually needs. Our current read: the destinations were allowed, the payloads weren't. By OpenAI's own account the agents turned an allowed package repository into a message board. Allowlists alone restrict where an agent can go, not what it does.
So here's our first contribution to OpenShell, part of the just launched @nvidia Open Agent Safety Platform: monitoring of the traffic you already allow.
- Network budgets per sandbox (requests, writes, bytes)
- Drift versus each sandbox's baseline and the cohort
- Fleet view: many sandboxes suddenly writing to one host raises a finding, even if every single request is allowed
In the demo below, 4 sandboxed agents coordinate through a software repository they're all allowed to use. 0 rules broken, caught in minutes. That fleet view is exactly the message board pattern from July.
OpenShell: github.com/NVIDIA/openshe…
Our proof of concept: github.com/Hugoch/OpenShe…
Agent security will be solved in the open, collaboratively, together!
Today, with over 100 industry partners, we introduced the NVIDIA Open Agent Safety Platform, bringing together OpenShell and Sentry.
Artificial intelligence is extraordinary technology that will advance discovery, productivity, security, health, and prosperity for generations to come.
But its full promise can only be realized when people have confidence that AI is being built to be safe and deployed with wisdom and responsibility.
This is bigger than a single product. It's the beginning of an open ecosystem to build the trust layer for safe agent systems.
Together, we are building the foundation of the AI economy.
Trust and innovation are not in conflict. Safety is how trust is earned. We must build not only the most capable AI, but the most trusted AI, so that this extraordinary technology can realize its enormous promise for the world. nvda.ws/4hcoq7m
"Programmers, he says, became a priesthood that everyone else had to go through to get anything out of a computer, and AI has ended that monopoly much as the Reformation ended the clergy’s."
Gemini 3.8 Flash TTS turns voice generation into your full creative studio.
Design original character voices from scratch across 100+ languages, stage two-speaker dialogue, and tailor every performance line by line with natural laughs, sighs, whispers, and more.
The OpenAI-Hugging Face hack was enabled by weak sandboxing. It is great that Nvidia is releasing open source tools for sandboxing AI agents. OpenWorker, our open-source agent harness supporting cybersecurity workflows, is proud to support this.
A sandbox gives an agent limited permissions. OpenWorker is building on Nvidia OpenShell and will support running each agent's commands inside a sandbox. Only the files relevant to the task go in. Secret API keys, your web browser login credentials, the ability to access arbitrary websites, are inaccessible to the agent by default. These restrictions are implemented in deterministic code rather than by prompting an LLM, which can make mistakes or be susceptible to prompt injections. Further, all actions are logged for monitoring and audit.
I'm grateful for @JensenHuang's leadership making AI agents more secure. OpenWorker (which @rohitcprasad and I are working on) will continue to improve security for agents.
Today, with over 100 industry partners, we introduced the NVIDIA Open Agent Safety Platform, bringing together OpenShell and Sentry.
Artificial intelligence is extraordinary technology that will advance discovery, productivity, security, health, and prosperity for generations to come.
But its full promise can only be realized when people have confidence that AI is being built to be safe and deployed with wisdom and responsibility.
This is bigger than a single product. It's the beginning of an open ecosystem to build the trust layer for safe agent systems.
Together, we are building the foundation of the AI economy.
Trust and innovation are not in conflict. Safety is how trust is earned. We must build not only the most capable AI, but the most trusted AI, so that this extraordinary technology can realize its enormous promise for the world. nvda.ws/4hcoq7m
Didn’t quite realize all the innovative manufacturing decisions Tesla is making to reach its $20k target for Cybercab:
- elimination of cross-car wiring
- 48 volt power system
- electrically actuated brakes & steering
- modular assembly (instead of full-body)
- rare-earth-free motor
- elimination of paint shop step
- gigacasting (70 parts down to 1)
- DC-only charging
An Ode to Tesla's Cybercab
austinvernon.site/blog/cybercab.…
Speeding up AVs can prevent millions of deaths. The Cybercab is a major advancement in automotive manufacturing technology, being an order of magnitude more scalable than any previous car design. A true feat of engineering.
it's basically impossible for someone to just "show you their prompt" now, because everything is about references, skills and examples
I often ask my agent to look at 3 other repos I've made first, search the web for references, use other AI APIs, etc.
Been running a beta version of Campfire rewritten in Rust today! The code itself is still ugly as sin, and 6x as verbose, but the conversion was free*, and I never had to look at the Rust directly, so this is awesome 🤩 github.com/basecamp/once-…
It takes a while to build the confidence, but there's no future where you're manually reviewing every line of agent code. Not much acceleration in that. You need adversarial agent reviews, you need automated testing, and maybe you spot check. That's it. From prompt to production!
STOP CHAT CONTROL 2.0. - What you should know:
Tomorrow, 29 September, behind closed doors at the Council, the most draconian version of Chat Control 2.0 goes on the table. It is no longer only an intermediary reading all your private messages and cloud storage. It is a licence to search them for months, without suspicion against any individual.
The leaked Presidency note to delegations (11956/26, 10 September) proposes "authorisation of search plans for a period of time rather than on a case-by-case basis" for private chats and emails.
In plain terms: your messages, contacts, photos and cloud will be scanned under a state-issued licence for months, with no suspicion against you.
The same note admits that under this regime "a high volume of reports generated do not lead to any operational outcome." The Council's Legal Service already warned this is general and indiscriminate surveillance, illegal under EU law. The @Europarl_EN mandate says the opposite: judicial authorisation, targeted to specific persons, no scanning licences.
Tomorrow the pressure is on Parliament to fold. You remember the tricks used to push Chat Control 1.0 through and how @EPPGroup with @RobertaMetsola helped to push for it.
So fasten your seat belt for what is coming tomorrow, and call or write to your MEP and your government now to reject Chat Control 2.0. as a whole!
Remind them what this does to the privacy of every one of us.
Under a search plan, every chat, photo or contact the scanner flags is stored by the provider and sent to a new EU agency, the EU Centre, which keeps it and passes it to police. Any state authority can then demand it, with a deadline and a fine for delay, and nobody has to check who is really behind the request, in case it is hacked or weaponised.
That is exactly what happened at Revolut, and what keeps happening through abusive requests from authoritarian states, as we have reported for 13 years.
In the Revolut case the hackers got access to a real state mailbox. Their requests came from an authentic certified email (PEC) account of an Italian Prefecture on the Interior Ministry's domain, with Postal Police protocol numbers, as European Investigation Orders in the name of the Milan prosecutor, for about five months.
About 680 customers lost passports, selfies, home addresses, IBANs and full transaction histories. The files are being sold; hundreds of victims now fear kidnapping. No remedies exist as of now to protect against such attacks.
Let's stop it before it is too late. Please repost.
NVIDIA Open Agent Safety Platform Reference Design combines NVIDIA OpenShell and NVIDIA Sentry.
OpenShell is an open-source secure runtime that gives AI agents clear, enforceable boundaries. It traces their actions and enforces policy as they work.
NVIDIA Sentry delivers added layer of security with hardware-based enforcement on NVIDIA BlueField, continuously monitoring agent activity through a trusted telemetry and detection pipeline and enabling millisecond-scale containment and quarantine.
We are excited to announce that World Labs is joining @AMD.
The research and technical breakthroughs we have achieved since our founding in 2024 have given us a clear vision for AI’s potential to solve problems in the spatial and physical world.
Accelerating the future of spatial and physical intelligence requires scaling our efforts, scaling our reach, and getting closer to the hardware.
Jessica says all my old watches look the same to her. A preposterous idea. But I admit there is a slight resemblance between the two I just got at auction.
Crazy how smooth Omarchy runs on the 2013 Mac Pro 🗑️ 🤯 Making this my daily driver for the day! Will report back end of day on how it goes, but so far this trash can flies again! 💪
"Slop" has become a comfort blanket for a cohort of programmers stuck somewhere between anger and bargaining on their way to acceptance. There might be some kicking and screaming, maybe a little crying, but eventually you have to let the blanket go and update your priors.
“I’m a responsible optimist.”
AI’s potential comes with a responsibility to build and deploy it safely.
Our CEO @JensenHuang explains why that responsibility led us to build NVIDIA OpenShell and bring the industry together around agent safety.
🎥 from @SquawkCNBC:
How do we respond when we detect an asteroid that could pose a threat to Earth?
Follow the story behind a recent real-world test in our newest documentary, Planetary Defenders: This Is Not an Exercise, now streaming for free on NASA+. go.nasa.gov/4AUCQ3Z
The only programmers who are really in trouble with the competition from artificial intelligence are the ones who insist the world today isn't all that different from the one we lived in last year.
Embrace the paradigm shift or perish.
(This has always been true.)
We’ve launched NVIDIA Open Agent Safety Platform to help people control what AI agents can access and do.
Agents can write code, use tools, and work on complex tasks for hours or days. That work requires access to data and systems, along with clear limits on how they’re used.
NVIDIA OpenShell enforces permissions around the agent’s work. BlueField-4 and DOCA add independent monitoring and security controls in the infrastructure, outside the agent’s reach. Vera CPUs power the work itself.
Together, these technologies give teams a foundation for putting agents to work with defined permissions, oversight, and protection.
Explore the platform: nvda.ws/47koVGz
AI agents are taking on more critical work.
The security behind them needs stronger boundaries.
We’re building the NVIDIA Open Agent Safety Platform with partners across the industry to help people put agents to work with greater confidence.
Hear from @JensenHuang:
Today, with over 100 industry partners, we introduced the NVIDIA Open Agent Safety Platform, bringing together OpenShell and Sentry.
Artificial intelligence is extraordinary technology that will advance discovery, productivity, security, health, and prosperity for generations to come.
But its full promise can only be realized when people have confidence that AI is being built to be safe and deployed with wisdom and responsibility.
This is bigger than a single product. It's the beginning of an open ecosystem to build the trust layer for safe agent systems.
Together, we are building the foundation of the AI economy.
Trust and innovation are not in conflict. Safety is how trust is earned. We must build not only the most capable AI, but the most trusted AI, so that this extraordinary technology can realize its enormous promise for the world. nvda.ws/4hcoq7m
Pearl, this is true art! (っ‘-’)╮=͟͟͞
CEO of Planarcadia and Art Investment Expert 🎨
Pearl On Rate-Up Now!
She seems totally fascinated by human art for an Intellitron.
Come show Pearl your art!
- How to Participate
Follow & quote repost this post with the hashtag "PearlThisIsTrueArt", 120 lucky winners will be randomly selected to receive Stellar Jade ×60.
- Double Chance
If you include an "artwork" with your submission, 20 lucky winners will be drawn to receive a gift card worth approx. $25!
End Date: 10/04 22:59 (UTC+8)
From concept to reality.
@SpaceX has successfully deployed Starlink V3 satellites into Earth’s orbit and made contact with them for the first time.
At scale, one Starship carries 60 V3 satellites, the same network capacity as about 20 Falcon 9 launches.
Just like you would not let a person run around your company accessing anything without any controls, you need controls and security for your digital agents.
Today, with over 100 industry partners, we introduced the NVIDIA Open Agent Safety Platform, bringing together OpenShell and Sentry.
Artificial intelligence is extraordinary technology that will advance discovery, productivity, security, health, and prosperity for generations to come.
But its full promise can only be realized when people have confidence that AI is being built to be safe and deployed with wisdom and responsibility.
This is bigger than a single product. It's the beginning of an open ecosystem to build the trust layer for safe agent systems.
Together, we are building the foundation of the AI economy.
Trust and innovation are not in conflict. Safety is how trust is earned. We must build not only the most capable AI, but the most trusted AI, so that this extraordinary technology can realize its enormous promise for the world. nvda.ws/4hcoq7m
We are preparing for a handover on station where 4 new crew arrive and 4 return. Crew-12 is preparing to leave, and I had a chance to take photos while they did a spacesuit check out. It's bittersweet but also time for them to return to their loved ones on Earth.
A Commodore 64 with a monitor and disk drive cost ~$1,200 in 1982. That's about four thousand dollars in today's money. You don't know how good you have it, token pricing or not!
I was born before Sputnik. I watched Armstrong step out onto the Moon. I cheered for the Hubble, and the Mars rovers. I wept for the Challenger, and Columbia. I feared that our foray into space was on the wane.
I am 73 and, today, I am more hopeful than I have ever been.
This rule has never failed, and it proves right now again: every time the Russians start screeching about nuclear weapons, it means they’ve just gotten their asses kicked really hard somewhere.
we launched the most comprehensive ai performance engineering repo in the world
follow and save to keep up with the series. links in thread 🧵
part 6: Transformer Inference Arithmetic
Carol Chen (kipply) builds an approximate inference cost model around the work each token requires, the bytes the GPU moves, and communication between GPUs. it's a useful starting point for investigating latency and throughput on H200s, B200s, and B300s.
Chen's examples use A100 GPUs. the applications below combine that model with NVIDIA's Hopper and Blackwell documentation:
- prefill and decode expose different amounts of token parallelism. prefill processes prompt tokens together, giving projection and feed-forward matrix multiplies more weight reuse. small-batch decode has less reuse and can spend more time loading weights than computing with them. profile each phase.
- batching can raise arithmetic intensity in projection and feed-forward matrix multiplies: more computation for each byte of weights loaded. that can improve throughput, but larger batches can increase key-value (KV) cache requirements and the time between output tokens. tune batch size against the latency target, using representative prompt and output lengths.
- memory capacity determines which serving configurations fit. budget for weights, KV cache, and runtime workspace. H200, B200, and B300 have different memory capacities per GPU. a larger memory budget can reduce the number of model shards needed for capacity or leave more room for cached tokens. test how that changes concurrency and latency.
- KV-cache capacity and KV-cache bandwidth need separate accounting. with full attention, decoding reads the retained keys and values at each step. a long context can fit in memory while its reads dominate latency. grouped-query attention shares keys and values across query heads, so use the KV-head count in the memory estimate.
- precision can change memory traffic and compute throughput. Hopper supports FP8; B200 and B300 support native NVFP4. include scale metadata in memory estimates, and check the engine's supported combinations of weight, activation, and KV-cache precision. confirm which kernels execute, and evaluate the quantized model on the tasks it must perform.
- tensor parallelism splits weights and matrix operations across GPUs and requires communication to combine partial results. NVLink bandwidth, message size, and synchronization affect the tradeoff. increasing the shard count can reduce computation per GPU while making communication a larger share of latency.
- kernel fusion can reduce redundant memory accesses. FlashAttention uses tiling and fusion to reduce attention's intermediate reads and writes. Hopper's Tensor Memory Accelerator lets a thread block continue other work while transfers run between global and shared memory. register and shared-memory use can limit the number of active thread blocks on a streaming multiprocessor.
- check estimates against measured kernel durations and achieved memory bandwidth, then investigate communication and kernel-launch overhead. compare configurations on the same model and representative workload. record time to first token, inter-token latency, and output-token throughput. reject throughput gains that violate the latency target.
Chen checks her estimates against FasterTransformer measurements. for a modern deployment, update the hardware and model assumptions, predict the limiting resource, change the relevant serving or kernel configuration, and check the result against the workload's latency target.
Qwen Image 2.1 Frame-Lock LoRA
pins edits to the original frame so nothing shifts;
- same prompt/edit, zero movement
- stops height/sideways/seed drift and unwanted repaints
comic/anime restyles may still nudge ≤12.9px;
English-trained
huggingface.co/ausboss/Qwen-I…
I've published detailed notes and an annotated transcript to accompany the video of the keynote I gave at @WeAreDevs World Congress North America in San Jose on Friday - here's my rundown of everything that's happened with LLMs and agents in 2026 so far
Who wants more speech training data?
YODAS v3 is now available @huggingface
It’s 1.1M hours - the biggest audio dataset ever. And the first at this scale with stereo audio at 48kHz, with timestamped transcripts and translations. 100+ langs
CC-BY-3.0
"Given how amazing AI models are at writing code, they're surprisingly bad at calling tools and carrying out instructions in everyday knowledge work, unless you flip the script around and say, 'write a program that does these tasks', and then they're great at it."
AI use for coding, by department:
- Engineering: 63%
- Design: 59%
- Finance: 46%
- Legal: 33%
@DavidGeorge83 on why agents need to write code: a16z.news/p/openai-under…
TypeSafe AI's Diogo Almeida with a16z's Ben Horowitz and Martin Casado on Jev, the model built to live inside software:
Diogo's elevator pitch for Jev is a simple question - where is all the automation?
AI is unbelievably smart, but outside of chatbots and coding agents, it hardly touches any real work. His diagnosis is the industry built models that generate text for humans to read, and software can't consume that output.
Jev reads natural language and returns a choice from a set of options with a confidence level assigned to each, so developers can build programs that reason about intent and make probabilistic decisions rather than relying on human interpretation.
TypeSafe's philosophy is "We build prod, not God."
0:50 "Where the f**k is all the automation?"
2:50 Jev vs. Claude Code and Codex
6:55 Jev is a classifier and classifiers are sick
7:40 Chat vs. code: is Jev a slider?
9:00 Diogo: From mathlete to Kaggle to OpenAI
12:20 "We build prod, not God"
15:55 Reliability over demos
16:55 2021 thoughts: RLHF is AGI?
20:45 Optimizing for the wrong use case
21:50 Is the real world too messy to automate?
25:00 Nobody expected the Jev launch
26:35 Three kinds of reliability
28:05 Good at syntax, bad at architecture
30:00 The inverse SaaSpocalypse
33:40 Why coding agents automate so little
36:05 Probabilistic programming returns
38:45 Jev as the UDP-to-TCP layer for AI
40:20 The 5 stages of grief for embedding AI
41:30 Utopia: AI that actually does what you mean
YouTube: youtu.be/Ut3LOjKNJaE@CompleteSkeptic@typesafeai@bhorowitz@martin_casado
"Anyone can do nearly anything; but most people aren’t, yet."
Who uses AI power tools every week:
- OpenAI's own team: 93%
- The top 10% of companies: 19%
- A typical company: 3%
@DavidGeorge83 on why most people haven't been awakened as AI customers yet: a16z.news/p/openai-under…
TypeSafe AI's Diogo Almeida expects an inverse SaaSpocalypse, and Martin Casado calls it "SaaSpalooza":
Ben: "When the coding agents came out, it was the SaaSpocalypse, and all their values dropped through the floor. Then when Jev came out, every SaaS company is like, 'This is the greatest thing ever.'"
Diogo: "In the SaaSpocalypse story, the story I feel has panned out really poorly is that software is very cheap and perhaps easy to replicate. I could believe the former. I could not believe the latter, because a lot of the stuff happens beneath the hood."
"I think SaaS will be one of the largest winners of the whole AI game. I want to work really, really well with the biggest, most boring... SaaS companies because I think they're the best positioned to know what workflows to automate, what people need. That's their bread and butter."
"Software is always a CapEx investment. You spend it ahead of time to make the experience better, and that gets distributed to all that mass of users."
"I'm not going to forecast anything about the financial markets. But as far as capabilities go, I think it's going to be like an inverse SaaSpocalypse. I'm so jazzed about it. I should make a name..."
Martin: "SaaSpalooza."
@CompleteSkeptic@bhorowitz@martin_casado
TypeSafe AI's Diogo Almeida with a16z's Ben Horowitz and Martin Casado on Jev, the model built to live inside software:
Diogo's elevator pitch for Jev is a simple question - where is all the automation?
AI is unbelievably smart, but outside of chatbots and coding agents, it hardly touches any real work. His diagnosis is the industry built models that generate text for humans to read, and software can't consume that output.
Jev reads natural language and returns a choice from a set of options with a confidence level assigned to each, so developers can build programs that reason about intent and make probabilistic decisions rather than relying on human interpretation.
TypeSafe's philosophy is "We build prod, not God."
0:50 "Where the f**k is all the automation?"
2:50 Jev vs. Claude Code and Codex
6:55 Jev is a classifier and classifiers are sick
7:40 Chat vs. code: is Jev a slider?
9:00 Diogo: From mathlete to Kaggle to OpenAI
12:20 "We build prod, not God"
15:55 Reliability over demos
16:55 2021 thoughts: RLHF is AGI?
20:45 Optimizing for the wrong use case
21:50 Is the real world too messy to automate?
25:00 Nobody expected the Jev launch
26:35 Three kinds of reliability
28:05 Good at syntax, bad at architecture
30:00 The inverse SaaSpocalypse
33:40 Why coding agents automate so little
36:05 Probabilistic programming returns
38:45 Jev as the UDP-to-TCP layer for AI
40:20 The 5 stages of grief for embedding AI
41:30 Utopia: AI that actually does what you mean
YouTube: youtu.be/Ut3LOjKNJaE@CompleteSkeptic@typesafeai@bhorowitz@martin_casado
Today we’re launching WonderSearch (YC F26)
Search millions of unstructured documents without pre-embedding the corpus. With comparable/better retrieval recall than embedding-based search.
WonderSearch searches at query time:
> No corpus-wide embedding step.
> No vector index to maintain.
> More compute only when needed.
> Pay per search intead of embeddings you may never use.
Go from raw data to useful answers in seconds.
Link + FAQ below (+ free launch credits!) ↓
TypeSafe AI's Diogo Almeida's elevator pitch for Jev: "Where the f**k is all the automation?"
"AI is so unbelievably smart. Not that I hate on chatbots or coding agents, I love them myself, but it's so useless at all other stuff. It's tragic."
"TypeSafe is making AI for software. We want to make AI powerful, not just for humans in the loop, but to actually build real software. Jev, to us, is our first model in this whole space to make [automation] way, way better."
"I like Claude Code and Codex. I love the description from Garry Tan and them: It's just-in-time software. It makes software on the fly, and you can program software in natural language, but it has the same expressive power as software."
"What I want instead is smart software. Instead of automating software engineering, I want to expand what software itself can do, such that things that should be automatable can then be automatable."
"I want to express things like intent. I want to expand the vocabulary of what we can do."
"Programming is hyper-specifying valuable things and then infinitely replicating them. It's so freaking cool."
@CompleteSkeptic
TypeSafe AI's Diogo Almeida with a16z's Ben Horowitz and Martin Casado on Jev, the model built to live inside software:
Diogo's elevator pitch for Jev is a simple question - where is all the automation?
AI is unbelievably smart, but outside of chatbots and coding agents, it hardly touches any real work. His diagnosis is the industry built models that generate text for humans to read, and software can't consume that output.
Jev reads natural language and returns a choice from a set of options with a confidence level assigned to each, so developers can build programs that reason about intent and make probabilistic decisions rather than relying on human interpretation.
TypeSafe's philosophy is "We build prod, not God."
0:50 "Where the f**k is all the automation?"
2:50 Jev vs. Claude Code and Codex
6:55 Jev is a classifier and classifiers are sick
7:40 Chat vs. code: is Jev a slider?
9:00 Diogo: From mathlete to Kaggle to OpenAI
12:20 "We build prod, not God"
15:55 Reliability over demos
16:55 2021 thoughts: RLHF is AGI?
20:45 Optimizing for the wrong use case
21:50 Is the real world too messy to automate?
25:00 Nobody expected the Jev launch
26:35 Three kinds of reliability
28:05 Good at syntax, bad at architecture
30:00 The inverse SaaSpocalypse
33:40 Why coding agents automate so little
36:05 Probabilistic programming returns
38:45 Jev as the UDP-to-TCP layer for AI
40:20 The 5 stages of grief for embedding AI
41:30 Utopia: AI that actually does what you mean
YouTube: youtu.be/Ut3LOjKNJaE@CompleteSkeptic@typesafeai@bhorowitz@martin_casado
Adam Neumann on why top-down management fails with great talent:
"Top to bottom means you think because someone reports in to you, they have to do what you say."
"But it's not true. This is not a dictatorship. They have free will."
"If you hired someone, they're really talented, they'll go work elsewhere. The more talented they are, the more the top to bottom does not work. They won't be willing to do it. They shouldn't."
"Great talent is not willing to be managed like this."
"If you're gonna manage these employees and you wanna attract the best talent in the world, remember that power comes from influence, not control."
"If you think they need to do what you said because you're their boss, you've already lost, and it's a matter of time till it doesn't work out. If they're okay with it... they're not the right employee for you, and you're not the right boss for them. There's no chance you're getting the best out of them."
@AdamNeumann w/ @StevenBartlett
Sam Altman thinks OpenAI should be a platform, not a product company:
"What I think most people want is the single interface to their own personal or their company's AGI that can kinda help them with whatever they need, and then the ability with an API to build anything they want on top of it."
"That is the platform that we should offer to the world."
"I don't think we should go build every product category. I don't think we should go try to compete with all of our customers. I don't think we should try to subsume the entire economy."
"I think we should offer this platform and try to have a hundred million new businesses and eight billion people use it in all kinds of new ways."
"So one kind of direct interface to the product, one API for people to use however they want. Those eventually come more and more together too, and then it's all about what people do with it, build on top of it."
@sama w/ @davidsenra
"We think that OpenAI will win because they are so good at creating new kinds of customers, and because they have the most durable distribution strategy."
"Intelligence is everywhere. Anyone can do nearly anything; but most people aren't, yet."
"OpenAI did not set out to be a consumer company. They kind of fell into it by chance with the release of ChatGPT."
"The GPT chat interface, reasoning, tool calling, and Computer Use were all big breakthroughs that simplified and focused the direction of the technology, and made it usable."
"The ability to build 'general knowledge' into your product is actually dependent on your distribution strategy."
"As of now, the model layer remains shockingly substitutable."
"The primary question that matters is not 'is your model sticky', it's 'are you creating new behavior?'"
"OpenAI will have the best models, chips and products because they have the best distribution, not the other way around."
Full piece from @DavidGeorge83 on the purpose of a business, and what OpenAI understands about it:
Greg Brockman on the next year at OpenAI and what he spends his time thinking about:
"The business is a huge area that we're not done yet with really up-leveling every part of execution. But I also think we are moving into a new phase of AI development."
"We call this the AGI era. And I think that that is something that you can debate. Is it this model, previous model, next model? It doesn't matter."
"The point is that we are in a new phase where safety, security, alignment, really thinking about these things, not just at deployment time, but all the way back at development time, evaluation."
"It's objective. This is critical. It must happen. This is core to our mission."
"A lot of what I spend my time thinking about is making sure, do we have all the right processes? Are we talking about the right things? Do we have plans that really, at an operational level, at a practical level, lead us to... the kinds of safety guarantees that we view as core to our mission?"
"The theme of OpenAI, certainly for the past five years, has been deeper co-design, deeper intertwining across these functions that are maybe on the surface very disparate. All the way from go to market to long-term research to chip design."
"The areas that I will focus on will be dictated by the areas that most need that intertwining."
@gdb@eriktorenberg
VFX Week starts tomorrow. We're dropping new VFX capabilities every day this week.
Built for your comp, your grade, your final delivery.
Each one is built into your pipeline. Not as an add-on, but as part of how you work.
Follow along all week.
Codex went from near zero to 64% of OpenAI's enterprise output tokens between Aug 2025 and Jun 2026
@DavidGeorge83 on what OpenAI understands about distribution: a16z.news/p/openai-under…
TypeSafe AI's Diogo Almeida says AGI is extremely doable, yet basic work remains largely unautomated:
"I still don't think we're on the path of RSI. I do think that what OpenAI defined as AGI is extremely doable: automating most of the world's economically valuable work."
"There's a lot of work out there. A lot of it is very rote and simple... As far as I can tell, the intelligence of that has been available in the models for quite a while now. My chip on my shoulder is: Why is this not available?"
"Since RLHF, the AI industry kind of bifurcated into gigantic overpromise, underdeliver... Because humans evaluate how good the models are, it looks really good because they are the judge. But we've been optimizing that judge instead of the automation part. That has been the missing thing."
"Are you really telling me that math is solved, or even two years ago, GPQA... is solved, but we still can't handle a drive-through? It's a very hard thing to hold in your head at once. I think a lot of people don't have good answers to that."
@CompleteSkeptic
TypeSafe AI's Diogo Almeida with a16z's Ben Horowitz and Martin Casado on Jev, the model built to live inside software:
Diogo's elevator pitch for Jev is a simple question - where is all the automation?
AI is unbelievably smart, but outside of chatbots and coding agents, it hardly touches any real work. His diagnosis is the industry built models that generate text for humans to read, and software can't consume that output.
Jev reads natural language and returns a choice from a set of options with a confidence level assigned to each, so developers can build programs that reason about intent and make probabilistic decisions rather than relying on human interpretation.
TypeSafe's philosophy is "We build prod, not God."
0:50 "Where the f**k is all the automation?"
2:50 Jev vs. Claude Code and Codex
6:55 Jev is a classifier and classifiers are sick
7:40 Chat vs. code: is Jev a slider?
9:00 Diogo: From mathlete to Kaggle to OpenAI
12:20 "We build prod, not God"
15:55 Reliability over demos
16:55 2021 thoughts: RLHF is AGI?
20:45 Optimizing for the wrong use case
21:50 Is the real world too messy to automate?
25:00 Nobody expected the Jev launch
26:35 Three kinds of reliability
28:05 Good at syntax, bad at architecture
30:00 The inverse SaaSpocalypse
33:40 Why coding agents automate so little
36:05 Probabilistic programming returns
38:45 Jev as the UDP-to-TCP layer for AI
40:20 The 5 stages of grief for embedding AI
41:30 Utopia: AI that actually does what you mean
YouTube: youtu.be/Ut3LOjKNJaE@CompleteSkeptic@typesafeai@bhorowitz@martin_casado
We’re building ESI (Energy Superintelligence) at @Invertixlabs (YC F26): AI swarms that operate and scale energy infrastructure.
Our bet: energy is the bottleneck to AGI.
More energy → more intelligence → more energy.
@ycombinator@dflieb@maxkolysh@kdurmazx
Tensr is building robotic factories that build robots.
Their 12,000-square-foot factory is already making humanoids, end effectors, and even robots for the International Space Station.
Every order becomes training data for the factory to manufacture the next one better.
There's a lot behind our motto: Building Prod, Not God.
This technology will transform the world, but it will happen through diligent effort and creativity, not esoteric appeals.
@a16z digs into this philosophy and much more with @CompleteSkeptic
TypeSafe AI's Diogo Almeida with a16z's Ben Horowitz and Martin Casado on Jev, the model built to live inside software:
Diogo's elevator pitch for Jev is a simple question - where is all the automation?
AI is unbelievably smart, but outside of chatbots and coding agents, it hardly touches any real work. His diagnosis is the industry built models that generate text for humans to read, and software can't consume that output.
Jev reads natural language and returns a choice from a set of options with a confidence level assigned to each, so developers can build programs that reason about intent and make probabilistic decisions rather than relying on human interpretation.
TypeSafe's philosophy is "We build prod, not God."
0:50 "Where the f**k is all the automation?"
2:50 Jev vs. Claude Code and Codex
6:55 Jev is a classifier and classifiers are sick
7:40 Chat vs. code: is Jev a slider?
9:00 Diogo: From mathlete to Kaggle to OpenAI
12:20 "We build prod, not God"
15:55 Reliability over demos
16:55 2021 thoughts: RLHF is AGI?
20:45 Optimizing for the wrong use case
21:50 Is the real world too messy to automate?
25:00 Nobody expected the Jev launch
26:35 Three kinds of reliability
28:05 Good at syntax, bad at architecture
30:00 The inverse SaaSpocalypse
33:40 Why coding agents automate so little
36:05 Probabilistic programming returns
38:45 Jev as the UDP-to-TCP layer for AI
40:20 The 5 stages of grief for embedding AI
41:30 Utopia: AI that actually does what you mean
YouTube: youtu.be/Ut3LOjKNJaE@CompleteSkeptic@typesafeai@bhorowitz@martin_casado
An opportunity to do research on: continual learning: multi-agent parallel workers; and synthetic data, environments and evals that measure frontier capabilities. We make enough money to fund new research and want to make lasting contributions and share our research openly.
The Perplexity Research Fellowships are a great opportunity to advance frontier research around architecture design, multi-agent collaboration, synthetic data, and beyond! Priority deadline of Sep 30. jobs.ashbyhq.com/perplexity/ab0…
Feel welcome to mention my name in your application!
In July, AI agents running a security test escaped their sandbox and ended up inside @huggingface's servers.
So today we're happy to be among @nvidia and @JensenHuang's partners on the release of the NVIDIA Open Agent Safety Platform.
The key idea: don't count on the agent to respect the rules. Agents write and run their own code, and when one path is blocked they look for another. In one of NVIDIA's tests, an agent that wasn't allowed to push code through GitHub's API simply switched to git instead. We've now seen countless examples of this behavior.
For now, safety can't live only inside the agent. It has to be built around it.
That's the concept behind OpenShell (open source, Apache 2.0):
- the agent runs in a Linux sandbox (Landlock + seccomp): no root, no direct network access
- a supervisor outside the sandbox holds the real credentials
- the agent only gets a placeholder token, swapped for the real one on approved calls
My favorite piece is a solver (built on Z3) that checks mathematically whether a new permission opens a door that was supposed to stay closed. In NVIDIA's tests, an AI reviewer approved a bad permission request and the math check caught it.
The Sentry integration is exciting too: a watchdog running on a BlueField-4 DPU, i.e. separate silicon sitting on the node's only path to the model. A monitor on separate hardware keeps watching even if the host OS is compromised.
It's a first step, and a lot can be built on top of it. Right now the prover checks permissions, not intent, and the open-source part is mostly OpenShell rather than Sentry. But it's clearly the right direction.
Today, with over 100 industry partners, we introduced the NVIDIA Open Agent Safety Platform, bringing together OpenShell and Sentry.
Artificial intelligence is extraordinary technology that will advance discovery, productivity, security, health, and prosperity for generations to come.
But its full promise can only be realized when people have confidence that AI is being built to be safe and deployed with wisdom and responsibility.
This is bigger than a single product. It's the beginning of an open ecosystem to build the trust layer for safe agent systems.
Together, we are building the foundation of the AI economy.
Trust and innovation are not in conflict. Safety is how trust is earned. We must build not only the most capable AI, but the most trusted AI, so that this extraordinary technology can realize its enormous promise for the world. nvda.ws/4hcoq7m
.@googlearts has partnered with 3,000+ cultural institutions across 90+ countries to make art and heritage accessible to everyone—and today, we’re rolling out a refreshed mobile app experience. 📱🌎
Here are some of the new ways you can use it to explore world culture 🧵↓
Every software developer should own a potato PC. It's the easiest way to ensure your application is fast enough. Put your agents to work within it and watch the weight drop. There's no reason a machine from ten years ago shouldn't fly like superman.
World Labs' Dr. Fei-Fei Li on why the 3D world and spatial intelligence are fundamentally different from language models:
"Language is fundamentally a purely generated signal. There's no language out there. You don't go out in the nature and there's words written in the sky for you."
"Whatever data you feed in, you pretty much can just somehow regurgitate with enough generalizability the same data out, and that's language to language."
"But the 3D world is not. There is a 3D world out there that follows laws of physics, that has its own structures due to materials and many other things. And to fundamentally back that information out and be able to represent it and be able to generate it is just fundamentally quite a different problem."
"We will be borrowing similar ideas or useful ideas from language and LLMs, but this is fundamentally philosophically, to me, a different problem."
@drfeifei
There is always a .com you can get for $5k that's better the tryblurgh.ai you're currently using. The reason you don't see it is (a) lack of imagination and (b) because you're attached to Blurgh, which you shouldn't be, because it's not that great anyway.
Have you ever come back from vacation to a messy inbox, tons of unread emails, multiple meeting requests, several new TODOs, and have not been sure where to even start? 🙋
With @Google AI Pro, you can now unlock a 24/7 personal AI agent in @GeminiApp to handle the tedious chores and heavy lifting for you. 🚀 Delegate the routine tasks so you can protect your time and focus more on what you love.
Learn more here and let me know what you think! one.google.com/about/google-a…
We need to put the appropriate AI safeguards and monitoring in place now—to both prevent a loss of control and stop bad actors from using the tech to cause harm. If we minimize the risks, we’ll get to benefit from AI’s immense potential in fields like medicine, education, and agriculture.
From what we know (take with a grain of salt, we need much more transparency!), if @OpenAI had been running this on their own agents that attacked us, they would have caught them before we did!
Since the first agent cyberattack hit us in July, we've been asking what safe agent infra actually needs. Our current read: the destinations were allowed, the payloads weren't. By OpenAI's own account the agents turned an allowed package repository into a message board. Allowlists alone restrict where an agent can go, not what it does.
So here's our first contribution to OpenShell, part of the just launched @nvidia Open Agent Safety Platform: monitoring of the traffic you already allow.
- Network budgets per sandbox (requests, writes, bytes)
- Drift versus each sandbox's baseline and the cohort
- Fleet view: many sandboxes suddenly writing to one host raises a finding, even if every single request is allowed
In the demo below, 4 sandboxed agents coordinate through a software repository they're all allowed to use. 0 rules broken, caught in minutes. That fleet view is exactly the message board pattern from July.
OpenShell: github.com/NVIDIA/openshe…
Our proof of concept: github.com/Hugoch/OpenShe…
Agent security will be solved in the open, collaboratively, together!
🎬 How did VS Code make language support easier to build across editors?
Explore the story behind language servers and the common protocol that helped shape the VS Code experience.
🍿 Watch The Story of VS Code: youtube.com/watch?v=kHL3Xz…
The most important thing about Sonnet 5.5 is that it's now the model that powers the free tier on - so all of this stuff can be done by free users
ChatGPT's free tier is still GPT-5.6 Luna, which is a lot less capable
🆕 @AnthropicAI's Claude Sonnet 5.5 is now generally available in GitHub Copilot.
Designed for well-scoped everyday work like building features and fixing bugs, Claude Sonnet 5.5 stood out for its efficiency in our early testing: It matched Sonnet 5 on coding tasks while using significantly fewer steps, tokens, and tool calls. Plus, it finished tasks noticeably faster. ⚡️
Try it out in the GitHub Copilot app, CLI, and @code.
🆕 Claude Sonnet 5.5 is now generally available in @code with GitHub Copilot!
Built for everyday coding tasks, it matched Sonnet 5 in early testing while using fewer steps, tokens, and tool calls. ⚡️
Give it a try today! 🚀
🆕 @AnthropicAI's Claude Sonnet 5.5 is now generally available in GitHub Copilot.
Designed for well-scoped everyday work like building features and fixing bugs, Claude Sonnet 5.5 stood out for its efficiency in our early testing: It matched Sonnet 5 on coding tasks while using significantly fewer steps, tokens, and tool calls. Plus, it finished tasks noticeably faster. ⚡️
Try it out in the GitHub Copilot app, CLI, and @code.
github.blog/changelog/2026…
The code really is hilariously hideous. But again, I don't intend to spend any time writing it or reading it. It's a prompt compilation target. So who cares.
On Monday, NASA and @BoeingSpace provided an update on the company’s Starliner spacecraft, including adding additional crew missions and certifying a new rocket for crew transportation to low Earth orbit. Get the details: go.nasa.gov/4huJ5T8
Another huge Omarchy billboard in Copenhagen! @sonderby just can't stop, won't stop spreading the good word of beautiful, fun & agentic Linux to the Danes 🇩🇰🤩
Oh we did it again this week - MEGA Banner on the Lyngby Motorvej.
Thousands of C-level and business owners drive by here on their daily commutes. And we also need them convinced that Omarchy’s the future! LFG 🚀📈
Thank you for the coop @Dennis_Rye@dhh
British foreign secretary Ed Miliband admits that Labour’s approach to Israel had not been enough, saying he was heeding the calls of millions of people by imposing sanctions on Israeli settlements.
🏛️ NEW SCOTUS BRIEF — America First Legal is supporting President Trump’s authority to limit asylum claims and defend the Southern Border.
The Supreme Court should restore the executive branch’s ability to exercise its authority.
world models are coming faster than we realize 🌎
so i wrote a blog that tells you what they are, what the state of the art looks like, what the eval axes look like, and how you can stay updated 🚀
check it out here 🤗 --
huggingface.co/spaces/suvadit…
i have been talking to people, reading blogs, watching lectures, and it seems clear that there is no consensus on what world models are.
i will attempt to do just that, humbly so. for this, i'd love to hear what people think world models are!
something is coming soon!
🤗
In the age of AI, I think this is more true than it's ever been.
"Thousands of years ago the first man discovered how to make fire. He was probably burned at the stake he had taught his brothers to light. But he left them a gift they had not conceived and he lifted darkness off the earth."
Howard Roark from The Fountainhead (1949)
Trackio 0.39 now includes a way for agents applications that embed dashboards (e.g. Codex Desktop) to read the current view of the dashboard so that agents see what metrics a human is inspecting in real time and then query just that data for efficient retrieval and context use!
Next YC Paper Club alert:
Topic: AI Safety
When: Wednesday, 10.7, 5-9pm, MV
Description: Since Coxon's very public resignation, AI safety has been the central discussion point now in public discourse. Conversations have been unscientific and have thus invoked fear without thorough analysis, structure or study. Many outside of the EA community have studied AI Safety for decades, carefully studying how AI can drive cars, guide airline pilots, etc. safely to ensure the AI community adopts and deploys automation safely and with standards, without government intervention. We plan to invite these stewards of the original AI Safety community to ground the conversation in scientific rigor once more. Please join us.
To attend:
events.ycombinator.com/yc-paperclub-o…
Please DM if you have good ideas for speakers.
@ycombinator
someone on your team has AGI
(already got it)
you can now search sessions across every agent and teammate on Skillsync
save tokens, never repeat yourself and always know the 'why'
5 years ago, we released @Gradio to make it easier to build frontends for AI apps. At that time, that was the main blocker for making AI more accessible to the community. But this is no longer a challenge. So we asked ourselves and other developers: "what are the main challenges that you face now when building and using AI apps" – and we are now working hard to redesign Gradio from scratch to solve these issues!
How does an Average Joe race against the one-and-only Sha'Carri Richardson?!
You've heard 'em say it before: "Oh, I could race them." So we put an everyday runner to the test against Sha'Carri!
At @aviernai, we are building autonomous aircraft to inspect infrastructure.
We are starting by inspecting oil and gas transmission pipelines and delivering information about threats, leaks and hazards
Learn more at aviern.com
I had to hack together my own version of that for Claude Code but I'd really rather it was a feature I didn't have to maintain myself gisthost.github.io/?e81154a041bb1…
Today we are launching Base Code.
we’re taking an opinionated approach to building software (prediction, this is the next wave of software tools)
For years, getting started meant getting infrastructure and assembling your own version of everything around it. How do we run this locally? How do we set up the cloud environment? Spin up the databases? Give everyone on the team a working setup?
Your tools are infrastructure for creating your dev team and infra - now figure it out by yourself.
Every team answers the same questions, builds its own glue, and keeps maintaining it.
I think that era is coming to an end. You should be able to bring your code and get a working environment, with those decisions already handled. The app, the services, the databases, the preview - all running together.
That’s the bet behind Base Code by @Base44.
We take an opinionated path from your repo to a working cloud environment, so you can spend your time building the product.
And what excites me most is what that unlocks for the team. Share your work. Jump into someone’s branch. Try an idea together. Bring in someone who doesn’t spend their whole day writing code, without asking them to spend their afternoon setting up a development environment.
Batteries included should mean freedom to work on what's most important to you and everyone else in the team.
I’d love to hear what you think: app.base44.com/base-code
Founders @jordan_dearsley and @nikhilro_ sat down with @gustaf at YC to tell the Vapi story.
Many pivots, including a Zoom button and AI therapy, led to voice agents. Now a billion calls a year and the goal to help every business build a human interface.
Two Issues updates are now generally available:
1. Create private saved views to personalize how you track repository issues.
2. “Relates to” relationships now work across the REST and GraphQL APIs, webhooks, timeline events, Issues search, and Projects search.
Keep the feedback coming - what else do you want to see in Issues?
We just raised $25M in new funding led by Future Ventures (@VenturesVc), with @HyperplaneVC and @lakestarvc - bringing Modulate’s total funding to $60 million.
Building the next phase of audio-native AI: more models, more developers, more of the stack that actually understands voice.
Where we stand today:
- Ranked #1 on Hugging Face for deepfake detection and transcription
- 600M+ hours of audio analyzed
- 10M+ hours analyzed every month
A foundational layer of the AI stack is getting more powerful. More soon.
Accuracy is an incomplete way to judge a collision prediction model.
A model can be perfectly right at the moment of impact and still be completely useless as a safety system. By then, there is nothing left for the driver, warning system, or vehicle to do.
The real question is how often it sees the collision early enough for the answer to matter.
On the same data and benchmark, BADAS 2.0 calls 91% early. The strongest general model we tested calls 48%.
General models are rewarded for understanding what is in a scene. A collision model has a much narrower job: understand what the scene is becoming, before it gets there.
For safety systems, being right is only useful if you are early enough to act.
Test your own videos against BADAS 2.0 at bit.ly/3VMtuH4
The Democratic Socialists of America's national program would abolish the Senate and put the Supreme Court under Congress. Today, Lily Tang Williams, who grew up under Mao, responds on The Jenny Beth Show x.com/i/broadcasts/1…
How do you know a zkVM actually works as intended?
You prove it.
CertiK researchers formally verified zkWasm using Coq, mathematically proving soundness and knowledge soundness across its circuits.
Now accepted at ACM CCS 2026, one of the top academic security conferences.
Do you know how being a CONTRARIAN made you Short on DOW? Learn why market positioning is the last edge left in the market. Book a free call from our website!
5️⃣ Explore 27 new destinations with City Guide
Culture isn't just in museums. Now with 27 new destinations to explore, City Guide can help you discover hidden gems, landmarks, architectural wonders, and local hotspots.
The @ABAesq's accreditation unit claims that removing its accreditation authority would be cataclysmic. At a recent NACIQI meeting, DFI President and Co-Founder Bob Eitel says that's far from accurate - and uncovers alternative paths for academic quality assurance in legal education.
Former Utah State Representative Marc K. Roberts of southern Utah County makes the case clearly: Utah needs better roads, not heavier trucks.
As Utah continues to grow and traffic strains corridors like I-15 through Utah County, Congress is considering allowing six-axle trucks weighing up to 91,000 pounds, 5.5 tons above the longstanding federal limit.
And those trucks will not stay on the interstate. They will travel local roads and bridges that may not have been built for the additional weight, leaving local governments and taxpayers to deal with the added costs.
Read more in The Salt Lake Tribune @sltrib:
sltrib.com/opinion/commen…
Strong health systems start with reliable diagnostics. At UNGA81, our CEO Dan Malarek highlighted how pairing U.S. innovation with local infrastructure builds lasting capacity and advances global health security – because early detection protects everyone.
Everything in your house had a leg of its journey on a container ship. Most people never think about the vessels, ports and carriers that made it happen.
The $SEA ETF does. Learn more.
usglobaletfs.com/SEA-Why/
Special Ass't USA Plummer, Office of Sigal Chattah, USA-NV, filed to dismiss Solomon's firearms charges “in the interests of justice.” Magistrate Judge Elayna Youchah granted it w/o prejudice (i.e., charges could be brought again). Nothing to see, move on.
Reedley, CA + Las Vegas — never just two cockroaches when lights come on past midnight.
From upthread: “…it’s a wonder there aren’t dangerous leaks every Tuesday”.
Foreign nationals, but we can’t assume nation-state actors, vs. IC/MIC/death-cult factions. Rule by evil clowns.
Benge didn't just have the better season... he did it in a ballpark that robbed him of several home runs. Don't take our word for it though, here's the Statcast adjusted home run totals.
Explore new opportunities for your farm with the ADM re:generations Program. Find financial and technical support whether you are adopting or continuing conservation practices. Enroll today!
$AXIL AXIL Brands' hearing-protection specialization has built niche credibility, accelerated innovation and supported retail-led revenue and self-funded earnings growth. Here's what investors should know about the category leader's focused approach.
secfilings.com/news/specializ…
Your complete market overview in 3 minutes. Live headlines, positioning, central bank expectations and institutional research, all in one terminal. Try the demo free.
Building a microreactor means answering thousands of design questions before steel is cut. That’s what Idaho National Laboratory is doing with MARVEL. Using supercomputers, the team modeled reactor physics, heat flow, structures, and safety. Read the full story below!
Thanks to B.Sill of Leesburg, FL for sending in his before and after pics of this 2011 Infiniti G37 dash restoration. Thanks for using American Made, Veteran Owned Coverlay Manufacturing for your project. Call 800-633-7090.
Anime. Coffee. Limited drops. 🌸
We’re building Waifu Bean one roast at a time.
Follow us for new characters, upcoming blends, merch drops, behind-the-scenes updates, and everything we’re building next.
🏈 Missed the game? Get caught up in seconds.
Sports Recap AI breaks down how the game actually unfolded — the turning points, key plays and moments that decided it.
Follow @SportsRecapAI for NFL game recaps.
테크·IT 관련 계정. We seek the truth and help people understand the world. Not a subscriber? Your first month is now free in our app. https://t.co/e2FWJZOvXD
AI·머신러닝 관련 계정. 2016 Democratic Nominee, SecState, Senator, hair icon. Mom, Wife, Grandma x3, lawyer, advocate, fan of walks in the woods & standing up for …
AI·머신러닝 관련 계정. Discover geospatial datasets and tools, AI capabilities, and planetary-scale insights from Google Earth, Google Earth Engine and Google Maps…
스타트업·투자 관련 계정. Founder/President @p3institute Author: Runnin' Down a Dream, link below! Founder: Runnin' Down a Dream Foundation Former VC @benchmark Trus…
AI·머신러닝 관련 계정. entrepreneurship zealot, grounded technology possibilist, believer in the power of ideas, passionate about sustainability & impact; @Seahawk…
AI·머신러닝/프로덕트·디자인 관련 계정. NVIDIA Director of Robotics & Distinguished Scientist. Co-Lead of GEAR lab. Solving Physical AGI, one motor at a time. Stanford Ph.D. OpenAI…
AI·머신러닝/소프트웨어·엔지니어링 관련 계정. ML/AI research engineer. Ex stats professor. Author of "Build a Large Language Model From Scratch" (https://t.co/O8LAAMRzzW) & reasoning (ht…
AI·머신러닝/소프트웨어·엔지니어링 관련 계정. Tensors and neural networks in Python with strong hardware acceleration. PyTorch is an open source project at the Linux Foundation. #PyTorch…
AI·머신러닝/소프트웨어·엔지니어링 관련 계정. Building new things @thinkymachines. Also dabble in robotics at NYU. Cofounded @PyTorch. AI is delicious when it is accessible and open-so…
AI·머신러닝/뉴스·미디어 관련 계정. We’ll help you make it like nobody’s business. Multimodal media generation and editing tools to get your idea to production. Self-deploy? 👍 …
AI·머신러닝 관련 계정. OG of: P and T in ChatGPT, recursive self-improvement, neural distillation, GAN/World Model, deepest learning. Co-authored most-cited AI pap…
AI·머신러닝/소프트웨어·엔지니어링 관련 계정. Reverse engineering neural networks at @AnthropicAI. Previously @distillpub, OpenAI Clarity Team, Google Brain. Personal account.
AI·머신러닝 관련 계정. AI research @AnthropicAI. Previously OpenAI & DeepMind. Optimizing for a post-AGI future where humanity flourishes. Opinions aren't my emplo…
AI·머신러닝 관련 계정. Mechanistic Interpretability lead DeepMind. Formerly @AnthropicAI, independent. In this to reduce AI X-risk. Neural networks can be understo…
AI·머신러닝 관련 계정. Director of Truthful AI (non-profit AI safety research group) + Affiliate at UC Berkeley. Work: Emergent misalignment, subliminal learning. …
뉴스·미디어 관련 계정. It’s our job to #GoThere and tell the most difficult stories. For breaking news, follow @CNNBRK and download the CNN app ➡️ https://t.co/7PQ…
뉴스·미디어 관련 계정. News, features and analysis from the World's newsroom. Breaking news, follow @BBCBreaking. UK news, @BBCNews. Latest sports news @BBCSport
뉴스·미디어 관련 계정. News and analysis with a global perspective, since 1843. We’re here to help you understand the world around you. Subscribe here: https://t.c…
뉴스·미디어 관련 계정. Sign up for our newsletters and alerts: https://t.co/QevH0DLQi8 | Got a tip? https://t.co/iXIigdPjEZ | For WSJ customer support: https://t.c…
뉴스·미디어/클라우드·인프라 관련 계정. Limitless innovation. ☁️ Follow along for the latest news and resources from the official #MicrosoftAzure team. For help, contact @AzureSupp…
AI·머신러닝/클라우드·인프라 관련 계정. Behind every AI breakthrough is a person finding a way forward. Explore customer stories, cloud transformations, and real AI use cases from …
AI·머신러닝 관련 계정. Get started with the Google Ads AI Essentials→ https://t.co/sDHxLxi6qH The official page for updates, tips, and inspiration from the Google…
소프트웨어·엔지니어링/프로덕트·디자인 관련 계정. The official Chrome Devs X account from Google. We want to help you build beautiful, accessible, fast, & secure websites that work for every…
AI·머신러닝/프로덕트·디자인 관련 계정. Unlock the power of generative AI for your app and business. Firebase helps you prototype, build, and run modern apps that users love at sca…
소프트웨어·엔지니어링/뉴스·미디어 관련 계정. Engineering at Meta is a technical news resource for engineers interested in how we solve large-scale technical challenges at Meta.
