Safety and security
OpenAI promises ChatGPT will step in earlier to prevent violence
OpenAI says it is expanding ChatGPT's protections against violence and taking on a new commitment to report on public safety. How all of this will work remains to be seen.
The essentials
- Claimed by one partyOpenAI says it is expanding ChatGPT's safeguards, meaning its protective barriers, to prevent violence (OpenAI News).
- Claimed by one partyOpenAI promises that ChatGPT will step in earlier to prevent violence (OpenAI News).
- Claimed by one partyOpenAI says it is taking on a new commitment to report on public safety (OpenAI News).
Why it matters
OpenAI is turning violence prevention into a public promise, which suggests it can be held to account. Without details, it will remain a statement of intent.
What we don't know
- Which specific measures are being expanded, and at what point ChatGPT will step in.
- What information OpenAI will provide, to whom and how often.
- Whether anyone outside the company will check that the measures work.
Sources
Was this story useful? Yes Not really