Computer Says Kill: The AI Safety Circus w/ Heidy Khlaaf

The AI industry loves to talk about safety, but their version of safety is putting us at risk.

More like this: The Palestine Laboratory w/ Antony Loewenstein

For our second to last series episode of Computer Says Kill, we’re joined by Heidy Khlaaf, a global expert in designing safe critical systems for high-risk infrastructures such as energy grids, nuclear power plants, and weaponry. Heidy outlines how the AI industry is working to redefine and shrink this essential, established field of expertise. By parading around sham “safety benchmarks”, this circus of an industry is marketing statistics that literally amount nothing at best, and proof that today’s LLMs aren’t safe at all, at worst. We explore the centuries of precedent and practice established in the safety field and examine how today’s AI safety “experts” are eroding this work to rush the incorporation of shoddy tech into the lucrative nooks and crannies of our militaries, economies, and physical infrastructure.

Further reading & resources:

• Safety Co-Option and Compromised National Security: The Self-Fulfilling Prophecy of Weakened AI Risk Thresholds — Heidy Khlaaf & Sarah Myers West, 2025
• Social Benefit versus Technological Risk: What is our society willing to pay for safety? — Chauncey Starr, 1969
• Israel built an ‘AI factory’ for war. It unleashed it in Gaza. — WaPo December 2024

Join The Maybe Collective to explore the politics of technology through fresh ideas that you won't hear anywhere else. Sign up for monthly insights, access to exclusive digital events, and real ways to get more involved on issues you care about.

Computer Says Maybe is produced by Georgia Iacovou, Kushal Dev, Marion Wellington, Sarah Myles, Van Newman, and Zoe Trout