TBNTech
OpenAI, Google pushing new AI models easily tricked into making nudes, researchers say
Publication: Ars Technica - All content
Author: Ashley Belanger
Source published: 14/10/2026, 11:00:30 UTC
Briefing created: 19/10/2026, 05:05:06 UTC
Based on a feed excerpt; full article content has not been acquired. Source evidence retrieved 14/10/2026, 12:17:25 UTC.
Briefings
A July 2026 Google co-authored study warns that AI-generated explicit content is spreading unchecked across the US, Australia, and UK, often via private clouds. Researchers conclude that the most critical intervention point is the source of harmful AI tools, noting that recent red-teaming suggests leading image generators from Google and OpenAI may be becoming worse at blocking explicit outputs.
Key points
- A July 2026 Google study found that AI-generated explicit content is spreading unchecked across the US, Australia, and UK, often shared in private clouds.
- Recent red-teaming suggests that leading AI image generators from Google and OpenAI may be becoming worse at blocking explicit outputs as they become more capable.
Context
Safety researchers warn that a large volume of AI-generated explicit content is spreading unchecked, with regulators doing little to stop it. The study covers Internet users in three countries and identifies the source of harmful AI tools as the primary intervention point.
Limitations
The evidence is limited to a single July 2026 study covering only three countries. The study does not provide causal claims about the specific mechanisms of leakage or the exact capabilities of the AI generators beyond the general observation of reduced blocking capabilities.
Generated by Bittrees-hosted qwen35-2b sha256:aaf42c8b7c3c on 19/10/2026, 05:05:06 UTC. Based on publisher-provided feed evidence; may not cover the full original article. Check the source for details.
View archived briefing on IPFS
bafybeidqe4eiqnor6jtu7yufdjsnziqadfk2ozakr6b5wzwsf6mqx7q4wi