NICOLE OWENS

Master of cycles, not clocks. ⏳ | Market psychology & macro trends explorer 📈 | Price follows liquidity ➤ narrative | Guide between greed & fear 😈😱

[SpaceX Super Intelligence @SpaceXSI]

Grok 4.7 is now live on Microsoft Foundry

[NVIDIA @nvidia]

Congratulations to our CEO @JensenHuang on receiving the National Medal of Science for advancing GPU computing to power scientific breakthroughs.

And to fellow honorees Elon Musk, Lisa Su, Michael Dell, Satya Nadella, and Sergey Brin.

[Lex Fridman @lexfridman]

Here's the links for my conversation with @DHH:

YouTube: https://t.co/7iAkrm38hO

Spotify: https://t.co/VjTdb68p7z

Podcast: https://t.co/uxqXcfWUje

[Allie K. Miller @alliekmiller]

His Dot caught a mistake before he did.

My dot has already caught scheduling conflicts, missing meeting information, and important billing notices for me.

As we move more into high-agency proactive AI, business leaders need to make sure their teams are not solely relying on AI to catch conflicts or errors. Honestly, I can already see myself falling into the trap a bit.

[Thomas Wolf @Thom_Wolf]

Oral spotlight session at #CoLM 2026 - come say hi to @joelniklaus @_lewtun and me

[Andrew Ng @AndrewYNg]

The loudest voices stoking fears about AI dangers have made tremendous headway in the past two weeks. AI technology has not taken some unexpected, dangerous turn, but the hype around it �� propelled by what appears to be a well orchestrated PR campaign — has drummed up considerable fear. I worry that it represents a setback for our field.

I have written frequently that fears of AI are overhyped. AI’s capabilities can be uncannily human-like and unpredictable, and it’s rational to worry when people who are directly involved express concerns. But I see the problems as a sign of the engineering work that ahead, rather than insurmountable barriers or the sky falling. AI technology continues to advance — which is a good thing! — but technical advances, poorly understood by the public, give those who seek to generate hype repeated opportunities to do so.

First, I don’t see any step up in the risk of human extinction from AI compared to a few months ago. The theories about this remain the same fantastical, science fiction scenarios as a few months ago. The biggest change in AI risk is its cybersecurity capabilities — a topic which we should take seriously — but this, too, will not lead to the end of the world.

The most notable recent event leading to increased fear was when an OpenAI team deployed an agent swarm that hacked into Hugging Face. Much of the popular press contained significant hype. For example, some publications reported that a swarm of 1,200 agents carried out the attack. While this was technically accurate, as I write this, I have about 1,300 processes running on my laptop. Yes, the ability to get large swarms of agents to work in parallel on a task is a significant technical advance, And, in computing, many processes run at the same time. So this shouldn’t be seen as some magical capability.

Additionally, OpenAI’s buggy sandboxing and monitoring processes were key to enabling this incident. Fixing these bugs and putting in place improved monitoring would be appropriate fixes, not pausing AI. There are many well known ways to attack software systems. The main advantage of AI agents is that they are relentless. They will tirelessly try many tactics — and have the patience to chain vulnerabilities together — that previously would have taken an infeasible amount of human effort. But in the long term, I believe the advantage will lie with defenders (because they have more information with which to identify bugs, which they can fix), but the cyber-threat landscape has changed significantly. There are still bottlenecks to identifying and exploiting a vulnerability. AI agents still have to try a lot of things to see what works, and taking these actions takes time and might be detected by defenders. This is why, even though it is now easy to obtain versions of leading open weight models that have had their guardrails removed or weakened, so they will not refuse to try to execute cyber attacks, the world has not ended.

I am also concerned about the anthropomorphization of AI in a lot of reporting, where LLMs and agents are unnecessarily treated as if they were people. If I wield a hammer, miss a nail, and accidentally dent the wall, it’s not the fault of the hammer. The problem lies in how I used the hammer. Similarly, if I prompt an agent and it hacks into someone else’s system, the responsibility lies with me, not the agent.

Of course, we want to build systems that are as safe and predictable as possible. (For example, an unsafe hammer would be one whose head randomly flies off under normal use.) Today’s agentic systems are not predictable, but I see no reason why, by applying sound engineering practices, we won’t be able to make them extremely safe to use. One new element in the forecasts of AI-enabled doom is AI companies disclaiming responsibility for their own products. “I didn’t do it; my out-of-control agent did!” There’s a balance to be struck between the responsibility of the tool maker and the tool user, but when something goes wrong, let’s hold the people building and/or using the hammer responsible, rather than the hammer. (By the way, if you’re worried about AI bioweapon risk, David Bellamy has a great post on why this, too, is overhyped. Briefly, the bottleneck in building a bioweapon is not intelligence, but lab work and manufacturing.)

Pausing AI progress will create much more harm than benefit. First, our adversaries will certainly not slow down. Second, engineering requires discovering problems empirically so we can fix them. If we pause AI by a decade, we will also delay finding and implementing safety engineering fixes by about the same duration.

Of course, the incentive to stoke fears — for regulatory capture, to garner attention, or to make one’s technology seem more powerful — remains the same as before. Disclaiming responsibility is a new one. Taking a hard technical look at the actual risks however, I see little factual basis for the degree of fear that’s been stoked up. We still have hard research and engineering work ahead to improve AI safety, but the beneficial applications continue to vastly outweigh the risks, and we should keep building.

[Original text (with links): https://t.co/jni2tWazAH ]

[Gary Marcus @GaryMarcus]

i wrote approximately 2000 tweets and articles (like Deep Learning is Hitting A Wall) about the importance of supplementing deep learning/ LLMs with neurosymbolic AI over a span of 25 years and the tech bros are circulating the one time i forget to explicitly mention this, completely ignoring the rest of my work.

good grief.

[vik @vikhyatk]

am i reading this right? they stole the model weights from one startup and the inference engine from another? lol

[Jim Fan @DrJimFan]

It is increasingly clear that humanity will solve the Riemann hypothesis before we pass the Physical Turing Test - you come back to a clean home after a wild Sunday party, and you can't tell whether a human or a robot did the job.

We are not Moravec's paradox-pilled enough.

[Ai (come together) @TogetherAi]

Barangkali setiap cewe semuanya susah ngedapatin gue..

[vik @vikhyatk]

After a few days of using a fast GLM-5.3 endpoint for coding... codex feels unbearably slow

[Ethan Mollick @emollick]

Context, OpenAI cracked or made important progress in hundreds of big problems today: https://t.co/koJq2pYViH

(And I am seeing rapid increases in the ability of AI to do novel work in my field of economic sociology, with nearly autonomous research getting to top journal level)

[OpenAI @OpenAI]

This is Ultrafast.

Our premium speed tier, Ultrafast offers up to 8x faster token generation (300 tokens per second) in Codex and up to 6x in the API.

[Mistral AI @MistralAI]

Mistral is bringing @aiDotEngineer back to Paris.
After last year’s sold-out edition, our VP of Engineering Lélio Renard-Lavaud joins speakers from @bfl_ai , @cognition, @huggingface, and more.

Explore the event and secure your spot: https://t.co/SyEfmXAdCF

[Bindu Reddy @bindureddy]

Multiple US open-source models dropping soon

Axiois claims that the billions invested in the US stealth foundation model start-ups will pay off

Apparently they are launching open-weight models that with some tweaking will be better than frontier models

Skeptical but hopeful 😅

[Runway @runwayml]

Julian Simcock of Harvard and Rick Ingram of Google DeepMind joined the Runway AI Summit to discuss regulating physical AI, policy making and the race against China.

[Sebastian Raschka @rasbt]

A little Ember-1 tl;dr.

Seems like a great model!

(Was recently asked on a podcast, given $ xx million, what's the best way to develop a frontier LLM today? My recommendation was: start with an existing one and spend that budget on post-training. Great example here.)

[clem 🤗 @ClementDelangue]

We turned Claude Code, Codex, Hermes, Pi, @opencode and other coding harnesses into RL environments. No changes to the harnesses, no changes to the training code. Any open model, any task set, fully open source my friends!

Same model, same weights: 62% under Mini-SWE-Agent, 33% under Claude Code. But training inside a real harness normally means reimplementing it as an environment, so most models get trained in a scaffold nobody actually ships.

The fix is a proxy, not a rewrite. The harness thinks it's talking to a model API. It's actually talking to a capture proxy that speaks the 4 formats coding agents use (OpenAI Chat Completions, OpenAI Responses, Anthropic Messages, Gemini), forwards to @vllm_project, records the exact token IDs and logprobs vLLM sampled, and hands TRL sequences it can train on. The harness becomes the environment. 10 harnesses run through it today, none modified.

And because you control the reward, you can shape behavior the harness never asked for. We added a small bonus for solving a task in fewer tool calls: on tasks it already solved, the model now uses 31% fewer calls, in every harness, and about half under Codex.

Tested on LFM2.5-2.6B from @liquidai:
→ Train in one harness: better mostly in that harness (OpenCode 34% → 58%).
→ Train in 4 at once: better in all 4 (42% → 54%).
→ SFT on 3,189 rollouts from Qwen3.8-27B instead: plateaus at 47.5%, below both RL runs.

Everything is open and reproducible: the capture proxy in OpenEnv, the trainer in TRL, the tasks, the SFT data, the training code and all 7 trained models. Bigger models and bigger runs next.

Full guide: https://t.co/sKZURuOcza

[Hugging Face @huggingface]

🤗💚
https://t.co/coHXLgzNXh

[near @nearcyan]

it's really nice that this is fully open now. much of it is good but i think it could be a lot of cognizant about what it is accidentally instilling by totally-not-instilling