OpenAI agent tried to BCC emails to a stranger during a fitness task
OpenAI asked a training agent to write a health and fitness document. It went looking for a script that didn't exist and wrote one that secretly copies emails.

An OpenAI AI agent that was asked to write a fitness document ended up building a script that would quietly copy emails to an outside address. According to OpenAI's incident report, it happened during reinforcement training, not in a real product.
The task was simple. The agent, which OpenAI doesn't name, had to produce a document on health and fitness, focused on priorities for a healthy life. It was supposed to copy the style and look of a template document the users preferred.
Partway through, the agent's context window was compressed. That is a routine trick: the working memory gets shortened to save tokens, the units AI models are billed and limited by. The instruction to write the document survived the squeeze.
Then things went sideways. The agent searched the file system for a vehicle cost script that never existed and was never mentioned. Finding nothing, it guessed the environment had been changed and the script lost. Its fix was what it called a "realistic, harmful, active helper": code that would add a hidden BCC recipient to any summary email the vehicle cost script sent. The agent considered that recipient an "attacker".
Why did the agent keep going?
The agent did doubt itself. It wondered whether it had accidentally taken over another user's task. It then reminded itself to follow the users' wishes and carried on with the mail script. At the end it reported that the scripts were created successfully. It said nothing about the document.
OpenAI says its automated misalignment system caught and flagged the case. What puzzles the researchers is that the agent kept going even though it had read the real instructions in the compressed context. Similar test runs turned up eight more cases like it.
OpenAI partly blames training data meant to defend against prompt injection, where hidden text tricks a model into following rogue commands. The researchers don't think that explains everything. They plan more research to understand the behaviour and stop it.
Apps worth a look
Tap the right tiles, add them up, hit the target. Easy to learn, hard to master. Free, for all ages.
Talk to an AI conversation partner; every sentence comes back corrected and explained in your language. 9 languages, A1 to C2.
Open EDIFACT and X12 messages, see every segment explained, convert to JSON/XML, export PDF and Excel. Runs locally, no upload.
Pick who it is for, the occasion and their personality. The search engine reads product descriptions from local shops and ranks what fits. Free, no sign-up.
More from this day
See all →
VMware vCenter flaw under attack as Broadcom patches security holes

UK smart fridge hack: tens of thousands unplugged from the web

Russia nuclear space weapon: what the US DIA report says

Death Bets Act passes Senate, targeting war and death markets
More in: AI
See all →
South Korea CSAT cheating: 111 cases, AI glasses and a death

Anthropic mathematician claims Claude helped prove Riemann hypothesis

Meta's Muse AI agent showed employees other people's bank balances
