# Alignment research should clarify what good behavior means

Markdown copy of https://realalpha.app/s/alignment-research-should-clarify-what-good-behavior-means/ on Real Alpha, the best minutes from long interviews with the people building AI. Cite and link the page, not this copy.

- Speaker: [Richard Ngo](https://realalpha.app/people/richard-ngo/) · Richard Ngo is an independent AI researcher; former OpenAI governance and DeepMind safety researcher. · Researches multi-agent intelligence after working on AI alignment at DeepMind and governance at OpenAI.
- Host: [Theo Jaffee](https://realalpha.app/people/theo-jaffee/) · Co-founder & host, MTS · One of the first members of a16z's new media team; earlier hosted the Emergent Ventures-backed Theo Jaffee Podcast.
- Host: [Sophia Dew](https://realalpha.app/people/sophia-dew/) · Host, MTS
- Show: MTS
- Episode: [AI Cannot Reliably Run a Company for Six Months | Richard Ngo](https://realalpha.app/v/YJjXjKJhUnI/), Nov 4, 2026
- Moment: 19:19 to 22:25 of the episode (3 min), playing on https://realalpha.app/s/alignment-research-should-clarify-what-good-behavior-means/; the episode from there: https://realalpha.app/v/YJjXjKJhUnI/?t=1159

## Quote

> “I **don't buy this idea** that you need to be in the labs using frontier models in order to do good research.”
> — [Richard Ngo](https://realalpha.app/people/richard-ngo/) at [20:04](https://realalpha.app/v/YJjXjKJhUnI/?t=1204) · MTS

## Takeaways

1. Research using frontier models may not generalize far.
2. Clarifying concepts can matter more than access to the latest models.
3. AI constitutions need clear arguments for why their rules are good.

## About this segment

Richard Ngo doubts frontier-model access is necessary for useful alignment work and calls for clearer concepts and arguments about desirable AI behavior. That could make the field’s research more robust and accountable.

From “AI Cannot Reliably Run a Company for Six Months | Richard Ngo”, Nov 4, 2026. [Watch the full episode](https://realalpha.app/v/YJjXjKJhUnI/)

## Transcript

Transcribed by machine (Whisper) from the episode’s audio, so a word may be off here and there.

[19:23](https://realalpha.app/v/YJjXjKJhUnI/?t=1163) can you go into a little more detail about which uh areas of research or research directions you think are least productive and most productive yeah um so my sense is that most things that people are doing which rely on using uh the very frontier models are not gonna generalize that far and so we've had enough time with llms right uh so far that if if you're doing research that is going to generalize from current models to the ones that exist in a year or two then probably it will also generalize from current open weight models to the ones that exist in a couple of years.

[20:08](https://realalpha.app/v/YJjXjKJhUnI/?t=1208) So generally, I don't buy this idea that you need to be in the labs using frontier models in order to do good research. The kinds of things that I am most excited about involve clarifying the sort of concepts that we're using. So basically, we have a range of concepts that were developed over a period of a few years, especially on LessWrong, that are now steering the decisions of very large companies and soon are going to be steering the decisions of major governments in pretty sensitive ways.

[20:49](https://realalpha.app/v/YJjXjKJhUnI/?t=1249) and mostly my sense is that the concept is just not up to scratch and that when people say things like a good example here might be the alignment faking paper from anthropics so in that paper they had um you know claude was demonstrating some behavior that depending on how you uh define alignment could be very aligned or very misaligned it's just like people could uh the the term could just be used either way, basically. And so my sense is that the sort of research that feels really robustly good right now is the kind that tries to specify, for example, what good behavior would look like. Claude has a constitution that's led it to behave pretty well in a bunch of ways.

[21:34](https://realalpha.app/v/YJjXjKJhUnI/?t=1294) But what's pretty interesting is that I don't know if anyone has actually argued for the ideas in that constitution i don't know if anyone's sort of written down like here is how we think um an ai should behave and here's why uh claude behaving like this at scale is actually going to be good for the world or if they've done this it's sort of more like a pr type thing i don't know if anyone's actually given a um like coherent clear argument for that and that's that's the kind of thing that uh you know many people could could have done it and uh it's just the difficult kind of research to do because tangling with these sort of philosophical concepts is not incentivized very much by most people. Right. So what do you think the actual plan should be for the AI safety