Technology · Open for opinions
Will OpenAI release an official tool against prompt injection or LLM vulnerabilities by 10/31/2026?
Consider security releases between 09/02 and 10/31 amid the debate about structural vulnerabilities in language models. 20 opinions, 50% Yes
- Yes — 50%
- No — 50%
Deadline: October 31, 2026 at 11:59 PM UTC · 20 opinions
Resolution rules
Consider security releases between 09/02 and 10/31 amid the debate about structural vulnerabilities in language models.
Source: www.technologyreview.com
Frequently asked questions about this event
How is this event resolved?
Janela de apuração: de 02/09/2026 (00:00, horário de Brasília) até 31/10/2026 (23:59). Resolve como 'Sim' se a OpenAI lançar publicamente ou disponibilizar em beta aberto uma ferramenta ou recurso de segurança dedicado à mitigação de prompt injection ou ataques adversariais em sua API ou ChatGPT até 31/10/2026, conforme publicado em openai.com/news ou openai.com/safety. Caso contrário, resolve como 'Não'.
How are points scored?
Points depend on the share of votes your side had when you gave your opinion. Being right earns 100 minus that share. Being wrong costs that share.
Right now Yes has 50% of the votes: an opinion on it that turns out right earns about 50 points, a wrong one costs about 50.