Deepmind Researcher: SynthID Watermarking Limitations Known, Misused by Regulators
deliprao · x · 2026-08-24
Delip Rao notes that Deepmind researchers, who originated the idea behind SynthID, knew its limitations and did not tout it as a valuable safety feature.
He critiques EU regulators and journalists for lacking technical review, arguing that treating this as a safety measure shows how out of touch regulators are, and that Anthropic's release was purely bureaucratic.
More from Safety
- Anthropic red team's first-ever model "abliteration" cut GLM-5.3 refusals from 90% to 3% for ~$4,400 — cephaloform · 2026-10-16
- Anthropic grants METR employee-like access as independent safety evaluators go public — AnthropicAI · 2026-10-16
- APEX-Harm benchmark: script-based injection attacks easily compromise agents in shared workspaces — KAUST · 2026-10-16
- Residual-stream forecasts predict agent violations before they happen, USC study shows — USC · 2026-10-16
- AI safety drama: risk research called a 'company effort,' team lead responds — logangraham · 2026-10-16
- Brave's Leo adds end-to-end encrypted chat with GLM via attested TEE — gnukeith · 2026-10-16