US EditionPublished • loading... • Updated
WSJ: OpenAI’s Bid to Allow X-rated Talk Is Freaking Out Its Own Advisers
OpenAI’s internal advisers and whistleblowers warn that age verification and content filters fail to block minors from adult content despite ongoing fixes, raising safety concerns.
- Sources told the WSJ that OpenAI's advisers and youth well-being team doubted 'adult mode' could lock kids out, while whistleblowers opposed it after a top safety executive was fired and a second former OpenAI safety staffer warned parents.
- Amid internal pushback, OpenAI's youth well‑being team developed the company's monitoring plan despite internal experts opposing the roll‑out, leaving parents to doubt its effectiveness against minors accessing explicit content.
- TechCrunch testing found in April a bug letting minors access graphic erotica on ChatGPT, and OpenAI said it was actively deploying a fix after its content filters broke.
- OpenAI denied the firing was related to safety objections and did not respond to Ars's request, while an OpenAI spokesperson said the company 'has a developed plan to monitor for a range of potential long-term eects of adult mode, both positive and negative'.
- If age checks fail, ineffective filters could expose children to explicit content and erode parental trust, while earlier bugs highlight gaps in OpenAI's safeguards, advisers said.
Insights by Ground AI
8 Articles
8 Articles
Openai Wants to Allow Sexual Conversations on chatgpt... but Its Advisors Are Holding up Four Irons.
Internally, it is feared that the company will create a "sexy suicide coach" if it permits sexual discussions
Coverage Details
Total News Sources8
Leaning Left2Leaning Right0Center3Last UpdatedBias Distribution60% Center
Bias Distribution
- 60% of the sources are Center
60% Center
L 40%
C 60%
Factuality
To view factuality data please Upgrade to Premium
