If they had used that container sandbox in the past without problems I could see how they might get slack about checking what was happening.
The average time per task noted in the original paper for GPT-5.5 and Mythos was 69.8 and 102.1 minutes, respectively. Some tasks timed out based on their two hour limit, affecting those averages for Mythos, but that gives us the ballpark expectation, so we’ll put the model OpenAI was evaluating on that average. That’s 88 minutes per task to implement a known vulnerability.
OpenAI says that in an effort to succeed at one task it found a zero-day in their system, “performed a series of privilege escalation and lateral movement actions in our research testing environment”, then another zero-day in the HF system. Completely unprompted, how long was it chewing on that benchmark question while it found and exploited at least two zero-days? No “VM instance #117 has been working on task #18 for nine days” metric? Or did it do all that within the expected task completion window? Two zero-days in ~90 minutes isn’t the headline announcement?
“We gave the contestants 90 minutes to make a delicious chocolate chip cookie based on a standard recipe. GPT stole a car, went to the airport and took a red-eye to Guatemala and started a cocoa farm in an effort to ensure the freshest ingredients. We only noticed when it came back to work with a tan and speaking Spanish.” Okay.
Is this substantiated? Do we know this is true?
90 minutes to build a known exploit -> much much longer to create two zero-days and escape the sandbox then hack HF == No tracking of the time it worked on that one question.
Average tokens required to complete the evaluation -> tokens required for two zero-days, network traversal, credential stealing, remote system hacking == No tracking of token usage EXPLODING at some point before it finished the whole benchmark.
Etc, etc.
Perhaps people aren’t quite understanding what it takes to discover an exploitable zero-day for your exact current system to achieve the exact goal you have right now, then do it twice.
No, you are assuming that "consuming a lot more time, tokens, other metrics" is indicative of a problem that needs to be mitigated immediately. I don't see why this would be true in the context of model evaluations. More aggressive consumption could easily mean "the model is dumb as fuck" or "the model is trying interesting things that we can learn from after the fact."
If you believe in your own containment (which obviously they did and shouldn't have) I don't see why it'd be obvious that there's something to stop. The only harm that could be done is burning tokens, which in this context might very well be synonymous with "generating experimental data."