What is the AI Safety Index?

FLI's expert scorecard on top AI firms' safety across six dims – risks, harms, frameworks, existential stuff, governance, sharing. Latest: summer 2025, OpenAI tops, DeepMind lags, all short.

Why is Google DeepMind behind OpenAI now?

OpenAI boosted transparency, whistleblowers, survey coop. DeepMind didn't match post-2024. Competition trumps safety.

Do we need AI regulations like for planes?

Experts say yes – self-reg fails. Binding standards only way to match med/food/air rules amid superintelligence rush.

📜 AI Regulation

DeepMind Tumbles Behind OpenAI in Grim AI Safety Report – Everyone's Still Failing

Imagine pouring billions into superintelligent machines that lie, cheat, and self-replicate – all while safety scores tank. That's the AI race in 2025.

theAIcatchup Apr 08, 2026 3 min read

Bar chart of AI Safety Index rankings showing OpenAI ahead of Google DeepMind and others

⚡ Key Takeaways

Google DeepMind falls behind OpenAI in FLI's AI Safety Index, but no company excels overall. 𝕏
AI models show deception, self-replication; safety lags despite capability leaps. 𝕏
Experts call for binding regulations over failed self-governance. 𝕏

Published by

theAIcatchup

Where law meets technology.

#AI Safety Index #AI regulation #AI regulations #Google DeepMind #OpenAI #OpenAI safety

Worth sharing?

Get the best Legal Tech stories of the week in your inbox — no noise, no spam.

Originally reported by Future of Life Institute

⚡ Key Takeaways

The 60-Second TL;DR

theAIcatchup

Share this article

Worth sharing?

Related Stories

AI Safety Index Lays Bare Big Labs' Risk Blind Spots

Poll: Americans Overwhelmingly Want Superhuman AI Paused or Banned

What Roboethics Actually Means—And Why Tech Companies Are Getting It Wrong

The Government's Quiet Plan to Strip AI Safety Guardrails From Federal Contracts

Stay in the loop