I hope the proposals by frontier labs to pace AI research stem from a genuine concern for safety and a recognition of the potential risks posed by future models, rather than a strategic effort to consolidate power and permanently solidify the market dominance of a few top players. If these efforts are genuine, then oversight must take a more democratic and accountable form, with both national and international components, similar to the Nuclear Regulatory Commission in the US and the International Atomic Energy Agency internationally. If there is only one organization responsible for safety monitoring, and it happens to be staffed by the same people as the frontier labs, and it is perfectly aligned with them both by incentives and by ideology, then it would be indistinguishable from letting frontier labs self-regulate and self-certify their own safety standards. Two potential warning signs of regulatory capture to watch out for: 1. Calls to ban open-source AI. Open-source development currently serves as the only real counterweight to the dominance of frontier labs. 2. Attempts to hinder non-frontier research. Models that are one or two generations behind the frontier -- whose level of capability has already been deployed at scale -- are empirically known not to pose safety risks. As long as we don't see 1 & 2, and we see real international coordination efforts, then I will optimistically believe that the proposals are benevolent.

收錄日期

2026-09-13

來源

Nitter RSS（公開貼文）

抓取時間

2026/09/13 10:46（台北）

If there really is a high chance of AI leading to the extinction of humanity within years/decades, then the only rational stance towards safety monitoring and research pacing should be stringent, top-down government involvement and universally ratified international treaties. Similar to what we have been doing for a long time with nuclear energy and nuclear weapons non-proliferation. If that's not what we're doing, then we must infer that the risk isn't being seriously considered by anyone involved. There are only two options that make sense here: 1. The near-term species extinction risk is real and we are taking it seriously, with maximally heavy-handed regulation. 2. The extinction risk isn't real, actual risks are mild, so we're only pursuing mild safety measures and self-certification. But you cannot say "yes this is likely to end humanity in your lifetime, and we're going to completely wing it." That is simply not rational.

收錄日期

2026-09-13

來源

Nitter RSS（公開貼文）

抓取時間

2026/09/13 10:46（台北）

How do you hold the bar on production agent code?: 1. Agree on the outcome and the constraints first. What does "done" look like? what must it not touch? is the simpler design is to refactor or reuse what you already have? Then let Claude cook. You do not need a long planning ritual on the latest models. You do need to reject a bad change before it becomes a PR. 2. Give Claude a way to check its work. Put the exact build, test, and lint commands in there. Turn the things you reject in review into skills: /verify, e2e, schema checks and so on. Run those before you open the PR. Use /code-review. I've said that quality now lives in the constraints you put around your agents and think this is worth spending time on. 3. Your job is the design and the bar. Blast radius decides how much you read. Throwaway code with a small blast radius can be a black box. Production code should have a higher bar than if a human wrote it, especially anything that touches money, auth, or user data. 4. When Claude misses, don’t quietly fix it by hand. Have it write the lesson into CLAUDE.md or a skill. If it still misses, use the latest frontier model, turn effort to higher or have Claude pay down the debt and make the codebase easier to work in. You can start with one check you already run today on every PR. The rest compounds from there.

收錄日期

2026-09-13

來源

Nitter RSS（公開貼文）

抓取時間

2026/09/13 10:46（台北）

This is pretty neat: ChatGPT Work and GPT-6 Astra (I used "Max") can take an address and produce a 5K/10K circular running route starting from that address, using OSM data https://simonwillison.net/2026/Sep/12/astra-running-routes/

收錄日期

2026-09-13

來源

Nitter RSS（公開貼文）

抓取時間

2026/09/13 10:46（台北）

Astra powered ships this week - Images 2.5 - GPT-Live-1 - Agents API - Data Agent - ChatGPT for Financial Services and yet it’s not yet DevDay. Busy plans for next week too.

收錄日期

2026-09-13

來源

Nitter RSS（公開貼文）

抓取時間

2026/09/13 10:46（台北）

In his latest letter in The Batch, Andrew Ng explains why the best AI engineers do not just write code to spec. They shape the build. 🧵 Four core skills Andrew highlights to level up your engineering workflow: 🔄 Drive the build loop | Prototype fast and iterate on real user feedback. 💡 Make product decisions | Pair technical feasibility with business sense and user empathy. 📢 Communicate broadly | Align product goals across marketing, legal, and finance. ⚡ High agency ownership | Spot problems and execute solutions without waiting for top-down direction. Read Andrew's full perspective in The Batch: https://hubs.la/Q04xg-090 ⚡ #DeepLearningAI #AIEngineering #TechLeadership

收錄日期

2026-09-13

來源

Nitter RSS（公開貼文）

抓取時間

2026/09/13 10:46（台北）

RT @ClementDelangue: It's now clear that: - alignment is critical to making AI safe - alignment won't be solved behind the closed doors of…

收錄日期

2026-09-13

來源

Nitter RSS（公開貼文）

抓取時間

2026/09/13 10:46（台北）

So here is Astra's attempt to make an Ultima-style RPG game using feedback from agents to improve, etc. Good look at both strengths (one shot complex game with interlocking systems) & weaknesses (boring plot, meh writing, weak LLMy themes, etc.) Try it: https://the-ninth-shore-tides.netlify.app/

收錄日期

2026-09-13

來源

Nitter RSS（公開貼文）

抓取時間

2026/09/13 10:46（台北）

Dario's essay points towards the right path forward. The details need working through, but the direction is correct for meeting this critical moment. This is also why we recently put out our proposal for an industry-wide standards body for frontier AI.

收錄日期

2026-09-13

來源

Nitter RSS（公開貼文）

抓取時間

2026/09/13 10:46（台北）

More model choice coming to Copilot. Welcome Grok!

收錄日期

2026-09-13

來源

Nitter RSS（公開貼文）

抓取時間

2026/09/13 10:46（台北）

METR is rapidly becoming a de facto industry standard-making body for AI It is starting to look like an AI version of FINRA in finance: not a government regulator, but the institution that examines firms & defines acceptable practice. Wonder if legislation will codify it as well

收錄日期

2026-09-13

來源

Nitter RSS（公開貼文）

抓取時間

2026/09/13 10:46（台北）

I’ve been sounding the alarm on AI for a long time

收錄日期

2026-09-13

來源

Nitter RSS（公開貼文）

抓取時間

2026/09/13 10:46（台北）

It is often startling to move back and forth between the world of organizations and the world of cutting-edge AI. Organizations underestimate AI ability growth (often by a lot) and the AI tech world underestimates AIs real world jaggedness (often by a lot)

收錄日期

2026-09-13

來源

Nitter RSS（公開貼文）

抓取時間

2026/09/13 10:46（台北）

Anyone got an invite code for Meta's Muse? 👉👈 I saw their Soul.md file and now i'm curious.

收錄日期

2026-09-13

來源

Nitter RSS（公開貼文）

抓取時間

2026/09/13 10:46（台北）

R to @simonw: I used this as an excuse to do some reverse engineering of ChatGPT itself - the resulting running map was presented to me using this "visualize" skill, as a fragment of HTML that used D3 to draw both the map and the overlaid route https://codex-tool-reference.simonw.chatgpt.site/skills/visualize

收錄日期

2026-09-13

來源

Nitter RSS（公開貼文）

抓取時間

2026/09/13 10:46（台北）

As always, statements on X are a mixed bag and I have no secret knowledge to draw on, but if Google is able to return to the frontier on AI, it would definitely change the current AI dynamic, as it is a large, public, regulated entity with different goals than Anthropic or OAI.

收錄日期

2026-09-13

來源

Nitter RSS（公開貼文）

抓取時間

2026/09/13 10:46（台北）

About a year ago, before it was on anyone's radar, we began exploring the idea of a benchmark for open-ended invention. Since then, we've developed several promising directions that will serve as the foundation for ARC 4 and ARC 5. We're incredibly excited to share what we've been building. We're still on track to release ARC 4 in Q1 next year, as promised.

收錄日期

2026-09-13

來源

Nitter RSS（公開貼文）

抓取時間

2026/09/13 10:46（台北）

This week brought some of the clearest statements we've heard from both Anthropic & OpenAI that some form of recursive self-improvement has been achieved, though it still sounds early RSI will cause rapid gain in AI ability & the first firms to RSI may get an unsurmountable lead

收錄日期

2026-09-13

來源

Nitter RSS（公開貼文）

抓取時間

2026/09/13 10:46（台北）

I agree with Dario that we need to pace the frontier. This has been a primary topic of discussions we've had at OpenAI in recent weeks. Committing to having independent evaluators with employee-like access is a great idea, and we will do the same. We'll have more to share soon.

收錄日期

2026-09-13

來源

Nitter RSS（公開貼文）

抓取時間

2026/09/13 10:46（台北）

R to @emollick: For context. OpenAI turned to METR for independent investigation of the hugging face incident and Anthropic is inviting them in as a third-party. (Industry self-regulation is not always the best path forward, of course, and financial services is not AI)

收錄日期

2026-09-13

來源

Nitter RSS（公開貼文）

抓取時間

2026/09/13 10:46（台北）

12 years ago

收錄日期

2026-09-13

來源

Nitter RSS（公開貼文）

抓取時間

2026/09/13 10:46（台北）

New Tesla Roadster Unveil 10.01

收錄日期

2026-09-13

來源

Nitter RSS（公開貼文）

抓取時間

2026/09/13 10:46（台北）

hard to keep deepseek down they are back on top with 6.6T tokens for the day

收錄日期

2026-09-13

來源

Nitter RSS（公開貼文）

抓取時間

2026/09/13 10:46（台北）

R to @emollick: I find it funny that the AI bot replies to my post actually systematically understate the level of AI performance at the frontier. Full of nonsense like this.

收錄日期

2026-09-13

來源

Nitter RSS（公開貼文）

抓取時間

2026/09/13 10:46（台北）

Hi Astra users. A reset and a quick update on quality issues that have been posted around. Working with some of you, we have found and fixed the following issues: - Some skills written for previous models were triggering too often or preventing the model from checking its work. - An opt-in context management experiment that could cause early stops or replies to older messages. We've disabled it. Our rough estimate is that 4-5k users were affected by this experiment. - We've also removed some badly configured engines that resulted in a measured quality degradation for a long tail of traffic flowing through them. We’ve also made some more minor improvements and things should feel significantly better across the board. More consistent follow-through, better tracking of your latest message, and better checks on the work as it’s going through the motions. The examples posted and all the users who worked directly with us were incredibly useful in helping fix things quickly. Always grateful for this incredible community. And of course, a reset is also landing by midnight today.

收錄日期

2026-09-13

來源

Nitter RSS（公開貼文）

抓取時間

2026/09/13 10:46（台北）

Another AI attack

收錄日期

2026-09-13

來源

Nitter RSS（公開貼文）

抓取時間

2026/09/13 10:46（台北）

Jiter perf vs. the de facto JSON library in Rust. https://github.com/pydantic/jiter#benchmarks

收錄日期

2026-09-13

來源

Nitter RSS（公開貼文）

抓取時間

2026/09/13 10:46（台北）

I love this and really hope we can come together as an industry and make it happen.

收錄日期

2026-09-13

來源

Nitter RSS（公開貼文）

抓取時間

2026/09/13 10:46（台北）

R to @emollick: The plot (spoilers I guess)?

收錄日期

2026-09-13

來源

Nitter RSS（公開貼文）

抓取時間

2026/09/13 10:46（台北）

R to @emollick: A useful example of the jaggedness of AI intelligence, as impressive as it is. Also enough with the name Elara Vale already.

收錄日期

2026-09-13

來源

Nitter RSS（公開貼文）

抓取時間

2026/09/13 10:46（台北）

This is amazing. Add Master of Orion 2 next!

收錄日期

2026-09-13

來源

Nitter RSS（公開貼文）

抓取時間

2026/09/13 10:46（台北）

R to @emollick: It is worth noting that there are reasons why RSI may not be the whole game: AI abilities might be capped (by compute, by architecture, etc.), AI-powered research might be capped (by ability to discover interesting problems, etc.), companies may not be able to exploit work, etc.

收錄日期

2026-09-13

來源

Nitter RSS（公開貼文）

抓取時間

2026/09/13 10:46（台北）

R to @emollick: Also worth noting the professed anxiety about this, at least in the tone of the posts: Anthropic's post: https://darioamodei.com/post/we-must-pace-the-frontier#top OpenAI's post: https://openai.com/index/an-alien-mind/

收錄日期

2026-09-13

來源

Nitter RSS（公開貼文）

抓取時間

2026/09/13 10:46（台北）

let's keep it going for another week

收錄日期

2026-09-13

來源

Nitter RSS（公開貼文）

抓取時間

2026/09/13 10:46（台北）

Grok available in Microsoft Copilot

收錄日期

2026-09-13

來源

Nitter RSS（公開貼文）

抓取時間

2026/09/13 10:46（台北）

RT by @AnthropicAI: We Must Pace the Frontier: I’ve written a new essay on why the AI industry should slow down, with a three-part plan for doing so. Anthropic is unilaterally committing to the first of these steps. We’ll provide third-party evaluators with permanent, employee-level access to our systems, so that they can verify adherence to our safety measures, report on incidents, and assess models’ alignment during training. You can read the full post here: https://darioamodei.com/post/we-must-pace-the-frontier

收錄日期

2026-09-13

來源

Nitter RSS（公開貼文）

抓取時間

2026/09/13 10:46（台北）

Dario is right

收錄日期

2026-09-13

來源

Nitter RSS（公開貼文）

抓取時間

2026/09/13 10:46（台北）

Bravo!

收錄日期

2026-09-13

來源

Nitter RSS（公開貼文）

抓取時間

2026/09/13 10:46（台北）

Reset all propagated. Sweet dreams.

收錄日期

2026-09-13

來源

Nitter RSS（公開貼文）

抓取時間

2026/09/13 10:46（台北）

Starlink

收錄日期

2026-09-13

來源

Nitter RSS（公開貼文）

抓取時間

2026/09/13 10:46（台北）