Safe execution environments are currently the bottleneck for autonomous AI development
Mostafa Dehghani · AI is Already Building AI — Google DeepMind’s Mostafa Dehghani · Apr 2, 2026 · at 12:07
Mostafa Dehghani is a research scientist at Google DeepMind. He discusses the difficulty of creating safe sandboxed environments inside Google where AI models can perform research engineering tasks autonomously.
0:00 / 0:29exact quote · 29.5s
▶ Watch the full episode on YouTube → 720p mp4 · rendered on demand · StarZero watermark
“Like in a safe setup, you know, where they can like, because right now we definitely, we don't, we're not confident about, you know, them doing the right things all the time and measuring like how much they can push and how long they can push a task is very difficult. And like connecting all these points into an environment that these models are operating and then get them run efficiently and like bringing diversity to eval is Definitely one of the bottlenecks of like making progress in this direction.”
quote is from the automated transcript, cleaned for reading: filler sounds and stutters are removed, nothing is rephrased. names can be misheard (the analysis reads context, assessments check outside sources). how →
More from Mostafa Dehghani
New pre-training techniques will drastically boost base AI model capabilities
“The way that we used to do pre-training, maybe, like, you know, like two, a year ago or two years ago maybe, like, you know, diminishing return is, like, obvious, but I can see how new ideas are bringing, like, you know, fresh, fresh energy into the pre-traini…”
Fully automated AI self-improvement will eliminate human bottlenecks and trigger breakthroughs
“The moment that we had this full automation, I would say we can close the loop of self-improvement and then it becomes the Like, you know, the problems become like, you know, mostly providing compute for these models to actually do what they want to do. And as…”
AI model progress will alternate between pre-training and post-training breakthroughs
“We're going to be having a bit of a swing back and forth between pre-training and post-training.”
Post-training techniques cannot compensate for a weak base AI model
“Pre-training is still the foundation and like, you can never post-train your way out of a week-based model.”
Video data conveys physical world knowledge to AI more efficiently than text
“So because of that, like picking up a lot of knowledge about the word through language is just not really efficient. I don't want to say that it's impossible, but it's not efficient, you know, like to learn about gravity. If you kind of like, you know, have yo…”
Demonstrating that image training lowers text perplexity remains extremely difficult
“So it turned out to be a really, really good model, but it was like really hard to see that. Wow. You know, I train on images and then like Text perplexity goes down. That was hard to see. You know, like the fact that, you know, you train in native model and i…”