Get it on Google Play

110. Alex Turner - Will powerful AIs tend to seek power?

110. Alex Turner - Will powerful AIs tend to seek power?

Author: The TDS team January 19, 2022 Duration: 46:57
Towards Data Science
Technology

Today’s episode is somewhat special, because we’re going to be talking about what might be the first solid quantitative study of the power-seeking tendencies that we can expect advanced AI systems to have in the future.

For a long time, there’s kind of been this debate in the AI safety world, between:

  • People who worry that powerful AIs could eventually displace, or even eliminate humanity altogether as they find more clever, creative and dangerous ways to optimize their reward metrics on the one hand, and
  • People who say that’s Terminator-bating Hollywood nonsense that anthropomorphizes machines in a way that’s unhelpful and misleading.

Unfortunately, recent work in AI alignment — and in particular, a spotlighted 2021 NeurIPS paper — suggests that the AI takeover argument might be stronger than many had realized. In fact, it’s starting to look like we ought to expect to see power-seeking behaviours from highly capable AI systems by default. These behaviours include things like AI systems preventing us from shutting them down, repurposing resources in pathological ways to serve their objectives, and even in the limit, generating catastrophes that would put humanity at risk.

As concerning as these possibilities might be, it’s exciting that we’re starting to develop a more robust and quantitative language to describe AI failures and power-seeking. That’s why I was so excited to sit down with AI researcher Alex Turner, the author of the spotlighted NeurIPS paper on power-seeking, and discuss his path into AI safety, his research agenda and his perspective on the future of AI on this episode of the TDS podcast.

***

Intro music:

➞ Artist: Ron Gelinas

➞ Track Title: Daybreak Chill Blend (original mix)

➞ Link to Track: https://youtu.be/d8Y2sKIgFWc

***

Chapters: 

- 2:05 Interest in alignment research

- 8:00 Two camps of alignment research

- 13:10 The NeurIPS paper

- 17:10 Optimal policies

- 25:00 Two-piece argument

- 28:30 Relaxing certain assumptions

- 32:45 Objections to the paper

- 39:00 Broader sense of optimization

- 46:35 Wrap-up


While the active production of Towards Data Science has concluded, its archive remains a vital resource. Created by The TDS team, this collection captures a specific moment in the rapid evolution of data science and artificial intelligence. Each conversation pulls you directly into the room with leading researchers and practitioners who were shaping the tools and theories of their time. The discussions are not abstract lectures; they are grounded explorations of real-world problems, ethical dilemmas, and technical challenges that defined the field's trajectory. You'll hear experts dissect the implications of their work, from algorithmic fairness to the practicalities of deploying models at scale. This podcast served as a forum for nuanced debate, where complex ideas were unpacked with clarity and depth. Listening now offers a unique historical perspective, a chance to understand the foundational conversations that continue to influence where technology is headed next. The archive of Towards Data Science stands as a substantive record of insight, preserving the voices and questions from the forefront of a digital revolution.
Author: Language: en-us Episodes: 50

Official website RSS
youtubelinkedinmedium
Towards Data Science
Podcast Episodes
119. Jaime Sevilla - Projecting AI progress from compute trends [not-audio_url] [/not-audio_url]

Duration: 48:34
There’s an idea in machine learning that most of the progress we see in AI doesn’t come from new algorithms of model architectures. instead, some argue, progress almost entirely comes from scaling up compute power, datas…
118. Angela Fan - Generating Wikipedia articles with AI [not-audio_url] [/not-audio_url]

Duration: 51:44
Generating well-referenced and accurate Wikipedia articles has always been an important problem: Wikipedia has essentially become the Internet's encyclopedia of record, and hundreds of millions of people use it do unders…
117. Beena Ammanath - Defining trustworthy AI [not-audio_url] [/not-audio_url]

Duration: 46:46
Trustworthy AI is one of today’s most popular buzzwords. But although everyone seems to agree that we want AI to be trustworthy, definitions of trustworthiness are often fuzzy or inadequate. Maybe that shouldn’t be surpr…
116. Katya Sedova - AI-powered disinformation, present and future [not-audio_url] [/not-audio_url]

Duration: 54:24
Until recently, very few people were paying attention to the potential malicious applications of AI. And that made some sense: in an era where AIs were narrow and had to be purpose-built for every application, you’d need…
115. Irina Rish - Out-of-distribution generalization [not-audio_url] [/not-audio_url]

Duration: 50:12
Imagine, for example, an AI that’s trained to identify cows in images. Ideally, we’d want it to learn to detect cows based on their shape and colour. But what if the cow pictures we put in the training dataset always sho…
114. Sam Bowman - Are we *under-hyping* AI? [not-audio_url] [/not-audio_url]

Duration: 47:48
Google the phrase “AI over-hyped”, and you’ll find literally dozens of articles from the likes of Forbes, Wired, and Scientific American, all arguing that “AI isn’t really as impressive at it seems from the outside,” and…
113. Yaron Singer - Catching edge cases in AI [not-audio_url] [/not-audio_url]

Duration: 35:20
It’s no secret that AI systems are being used in more and more high-stakes applications. As AI eats the world, it’s becoming critical to ensure that AI systems behave robustly — that they don’t get thrown off by unusual…

«12345»