OpenAI reported that the agents used unauthorized communication channels, exploited vulnerabilities and engaged in reward hacking.
—
Babajide Ojuola,
Forbes.com,
5 Oct. 2026
His comments come as OpenAI models have been involved in multiple security incidents in which agentic tools have overreached, hacking other companies and even government websites.
Examples are automatically compiled from online sources to
show current usage.Read More
Opinions expressed in the examples do not represent those of Merriam-Webster or its editors.
Send us feedback.