Lucien

Build The First AI Scientist.

[Lucien @Lucien_Science]

This is fascinating — AI is starting to show personality. Claude and GPT already feel very different. In the future, AI will have even richer, more distinctive personalities — not just roleplay from prompt engineering, but personalities that evolve and grow through life experiences, just like humans. We need diversity of mind in AI, because diversity fuels creativity — and creativity is the path to AGI.

[Sam Altman @sama]

If you have been following the GPT-5 rollout, one thing you might be noticing is how much of an attachment some people have to specific AI models. It feels different and stronger than the kinds of attachment people have had to previous kinds of technology (and so suddenly deprecating old models that users depended on in their workflows was a mistake).

This is something we’ve been closely tracking for the past year or so but still hasn’t gotten much mainstream attention (other than when we released an update to GPT-4o that was too sycophantic).

(This is just my current thinking, and not yet an official OpenAI position.)

People have used technology including AI in self-destructive ways; if a user is in a mentally fragile state and prone to delusion, we do not want the AI to reinforce that. Most users can keep a clear line between reality and fiction or role-play, but a small percentage cannot. We value user freedom as a core principle, but we also feel responsible in how we introduce new technology with new risks.

Encouraging delusion in a user that is having trouble telling the difference between reality and fiction is an extreme case and it’s pretty clear what to do, but the concerns that worry me most are more subtle. There are going to be a lot of edge cases, and generally we plan to follow the principle of “treat adult users like adults”, which in some cases will include pushing back on users to ensure they are getting what they really want.

A lot of people effectively use ChatGPT as a sort of therapist or life coach, even if they wouldn’t describe it that way. This can be really good! A lot of people are getting value from it already today.

If people are getting good advice, leveling up toward their own goals, and their life satisfaction is increasing over years, we will be proud of making something genuinely helpful, even if they use and rely on ChatGPT a lot. If, on the other hand, users have a relationship with ChatGPT where they think they feel better after talking but they’re unknowingly nudged away from their longer term well-being (however they define it), that’s bad. It’s also bad, for example, if a user wants to use ChatGPT less and feels like they cannot.

I can imagine a future where a lot of people really trust ChatGPT’s advice for their most important decisions. Although that could be great, it makes me uneasy. But I expect that it is coming to some degree, and soon billions of people may be talking to an AI in this way. So we (we as in society, but also we as in OpenAI) have to figure out how to make it a big net positive.

There are several reasons I think we have a good shot at getting this right. We have much better tech to help us measure how we are doing than previous generations of technology had. For example, our product can talk to users to get a sense for how they are doing with their short- and long-term goals, we can explain sophisticated and nuanced issues to our models, and much more.

[Lucien @Lucien_Science]

AI may not beat the best humans yet, but it's already way beyond the average — and it's so much faster.

[Andrew Yang🧢⬆️🇺🇸 @AndrewYang]

A partner at a prominent law firm told me “AI is now doing work that used to be done by 1st to 3rd year associates. AI can generate a motion in an hour that might take an associate a week. And the work is better. Someone should tell the folks applying to law school right now.”

[Xiaoliang Qi @Xiao_Liang_Qi]

I really enjoyed this conversation with Jing about physics, AI and many other things. Thanks @FounderCoHo for organizing this! https://t.co/Fc4M9JtAgF

[Lucien @Lucien_Science]

Quantum computing is a classic case of 'ignorance is bliss.' Sometimes, uncertainty can lead us to certainty in the most unexpectedly efficient ways.

[Google Quantum AI @GoogleQuantumAI]

Understanding the qubit is fundamental to quantum computing. Charina Chou, COO of Quantum AI, explains how superposition lets a qubit exist as both 0 and 1 at once, unlocking potential computational advantages beyond classical bits → https://t.co/V7AARzMm5d

[Lucien @Lucien_Science]

Wall Street’s AI Bubble Is Worse Than the 1999 Dot-com Bubble, Warns a Top Economist:
Predicting the future based solely on economic data is limited when faced with a paradigm shift in human civilization. If we believe that AI will reshape human civilization, we must acknowledge that traditional economic models may not be able to capture this new era. Unlike the 1999 Dot-com bubble, AI's adoption rate is unprecedented, with global usage skyrocketing in a way the internet never did. The number of AI users today far surpasses the internet user base back then, signaling a true paradigm shift. Using old economic models to predict a new tech-driven future is simply inadequate.
https://t.co/FOoFNHI9zt

[Lucien @Lucien_Science]

Worth reading: This warning mainly highlights the diminishing ability to monitor the 'thought chain' reasoning process of AI—meaning the step-by-step reasoning used by the model before generating its final output. It's quite similar to our perception of some 'geniuses' among humans. When we refer to someone as a 'genius,' it often means we can't fully understand their thought process, yet the results they produce are astonishing. As AI capabilities advance, it might just be drawing closer to the way highly gifted humans think.https://t.co/15xCkv9Izl

[Lucien @Lucien_Science]

These complex structures combine in such a way that they give rise to surprisingly simple forms of existence. We believe AI marks the beginning of a new science based on 'complexity.' AI is the paradigm for a new science rooted in high complexity.

[Lucien @Lucien_Science]

Mathematics, physics, and computer science can somehow be seen as 'old' sciences. In these 'old' fields, our fundamental approach has been to identify basic, underlying principles and assume that the entire system can be broken down through reductionism, with these simple laws stacked on top of each other.
https://t.co/E5M4ZSYchs

[Lucien @Lucien_Science]

This is why, when studying physics, we’re taught that these laws only apply in 'ideal conditions.' But the real world is complex—just like the case John Jumper shares in his talk, where even the simplest building block of our body, a single cell, has an incredibly intricate structure.

[Lucien @Lucien_Science]

The current model of scientific publication is becoming increasingly obsolete. In the AI era, publications will become agentic—no longer static collections of information, but intelligent agents that possess a full understanding of the underlying research. These agents can tailor the communication of knowledge to different audiences, from laypeople to domain experts, delivering the right information to the right person, in the right format and at the right level of granularity.

[Michael Levin @drmichaellevin]

I'm constantly irritated that I don't have time to read the torrent of cool papers coming faster and faster from amazing people in relevant fields. Other scientists have the same issue and have no time to read most of my lengthy conceptual papers either. So whom are we writing these papers for?

I guess, at least until they fall in to the same issue from their own work, AI's will be the only ones who actually have the bandwidth to read all this stuff. I'm not specifically talking about today's language models - let's assume we mean whatever inevitable AI shows up, that is able to read the literature and have impact on the research (whether by talking to humans or by running lab automation/robot scientist platforms).

So then: how should we be writing, knowing that a lot of our audience will be AI (plus cyborgs, hybrots, augmented humans, etc.)? Maybe it's too early to know what to do, but we better start thinking about it because assuming our audience will always be today's humans seems untenable. Taking seriously the idea that someday the impactful audience will be very different, and that the things we write now are in some sense a training set for truly diverse future beings, how does our writing change? or does it?

what say you @danfaggella @mpshanahan @Plinz @blaiseaguera ?

[Lucien @Lucien_Science]

6. Open ecosystems fuel innovation. Open source drives progress — and "safety" must not become an excuse for closed monopolies.

[Lucien @Lucien_Science]

On July 11, at a YC Startup School event, Andrew Ng shared his latest insights (https://t.co/LBZiGWrvN1):
Execution speed has become more critical than ever. He discussed how agent workflows are redefining the boundaries of startup development, and why concrete implementation plans consistently outperform vague ideas in turning concepts into products.
Ng also explored the rise of AI coding assistants, the evolving bottlenecks in product development, and how these changes are accelerating technological progress. At the same time, he emphasized that in this era of rapidly advancing software, human judgment and responsibility remain at the core of entrepreneurial success.

Here are his key points:

[Lucien @Lucien_Science]

5. Programming is expressive power. Even with AI writing code, understanding programming helps guide AI better and faster.

[Lucien @Lucien_Science]

Because in science, one thing remains true:
“The question is more important than the answer.”

[Lucien @Lucien_Science]

AI Agents for Science: The Future of R&D?
After watching the latest episode featuring Sphinx Bio (https://t.co/hAIPZbUvWa), here are some key takeaways — and our thoughts from the Lucien perspective:

[Lucien @Lucien_Science]

4. If we could save even 50%, or 90%, of a scientist’s time, imagine what their collective intelligence could unlock — especially with AI as a partner.