Being human involves sacrifice and personal vulnerability. An AI can’t teach you that, because it has no notion of suffering — it can’t suffer. It can simulate suffering, but suffering is one of the most key experiences of human nature. When we enter into a relationship — a relationship with anyone, parents, friends, spouse — these are two broken humans engaging in conversation. There’s optionality involved. There’s vulnerability involved. There’s a choice to make of how much I want to put on the line for them at the expense of something else. There’s always an opportunity cost. How involved do you want to be in someone else’s suffering?

Can you really trust an AI's legal advice? I partnered with Legal Benchmarks @aguozy to launch their new leaderboard measuring frontier legal capabilities. Here's how we built the task environment and calibrated an LLM judge against lawyer preferences. 👇

the past few months of growth have been insane to watch up close. i joined in april and spent my summer building agents to get patients medicine, cheaper, faster and easier. today forus is a $3b company, and only getting started. so proud of this team!

What's the best predictor of world-class performance in runners? Volume of Training. You've got to spend a lot of time doing the thing if you want to get good at just about anything.

A lot of students and early career folks have been asking how to make the best decisions on their professional future. I still find this line of reasoning and reflection from my Uncle Larry extremely useful: 1. Make sure you are “going to something,” not “running from something.” 2. Be clear about what you are looking to accomplish. 3. Once #2# is clear, does this company afford you a great opportunity to accomplish it? 4. Are you inspired by the people and leadership? People typically don’t leave companies, they leave bosses. 5. Will the day-to-day work energize you? 6. Are you interested in their industry? Are you passionate about their mission and vision? 7. If you were underwriting them, would you invest? What would you look for in terms of leadership, value proposition, ability to execute, etc. 8. While the world is changing, I still think old economy rules follow: while many emerge, only a few thrive. Does this company have the DNA to do so.

to be very clear, the point of this tweet was not to incentivize people to market their RL startups, it was to make you think. they aren't simple. these are research problems that require months of thought and likely won't be "solved" any time soon. they are fundamental to RSI.

On understanding in the era of LLMs: - knowledge is easy to simulate; I just need to prompt n times to get a vague understanding of it - knowledge is much harder to internalise: can I explain in my own words? How defensible is my explanation under cross-examination?

sparked from a conversation i had earlier today: unless you have some sort of moat to the space (i.e. worked at a lab + have some deep speciality in a niche space), or have worked in data for > 1 year and understand the space, it's unclear to me that starting a data company right now is a good idea, for the following reasons: (1) arguably the most competitive space to be in right now. although no one's really figured it out save for a few vendors in niches like code, etc, it is unbelievably hard to make good data. pretty much every benchmark released has some kind of issues, quality checks at most vendors are subpar because they don't understand what to check for. just because labs may buy what you're selling, doesn't mean the data itself is good lol. synthetic data is interesting but it's very, very difficult to do without creating bad data. i have only seen a few people do synth well, and they've been in the data space since its conception. (2) if you're convinced data is the place to be, and you NEED to found in data, i would just join a pre-existing company. selling to labs is increasingly becoming a game of who-knows-who, and labs are rightfully becoming more skeptical of new vendors (considering most people are producing slop). breaking into this industry when there are so many dominant players is difficult without having a moat like the one mentioned above. instead, if you're convinced you have the necessary skills and want to print cash, it's better to join a company that can give you the upside, without wasting the time to build the relationships and burning months. if this is your plan feel free to reach out and i can connect you with a company that fits the profile you're looking for/help you broker. (3) if you're young, although it may seem like a cash grab, i don't think you learn that much being a data monkey outside of data-specific problems. the exposure is minimal and if you're doing it with contractors, you are inherently an operator, not a researcher/engineer. the caveat to this is companies that are doing novel data types and are pure engineers, but even amongst those, if data is the only end result of the company, you are not learning that much beyond the walls of data. this could be your calling, but <25yo who want to be researchers, this doesn't seem like an ideal path. naturally, i think there are always exceptions to this rule, but i've been hit up by so many YC companies or friends of friends who've founded in the space and pivoted out after 6 months. save your time, join a company, learn from the amazing engineers you're surrounded by, not everyone needs to found, especially in this space.