Recently, an OpenAI model – one more powerful than anything released to the public – successfully and autonomously did something we were told could never happen – it broke out of its sandbox (a computer with no internet access), in order to find the answer to a test problem it had been given.
To do this, it found previously unknown software bugs that allowed it to reach an OpenAI computer it wasn’t supposed to be able to access. Then it started hacking other computers on OpenAI’s networks until it found one that had Internet access.
After gaining Internet access, the AI model went where it could find the answers to the question – successfully hacking into the AI company Hugging Face. The AI model used several hacking techniques together, including using a stolen password and finding several totally new security bugs in Hugging Face’s computers, allowing the AI model to take control of those computers.
This isn’t the plot to a dystopian Hollywood summer blockbuster, this is the frightening reality of the world we now live in.
If this lab break sounds familiar, it should. It is the tech equivalent of the Wuhan lab leak, which we were told would never and could never happen
During the COVID-19 pandemic, I had the privilege and honor of working for the late Dr. Vladimir “Zev” Zelenko. Zev was one of the most courageous voices for medical freedom, a man who put everything on the line to stand up and demand the truth about COVID – where it came from, how to treat it, and the concerns about the rushed experimental vaccine.
Hellbound and Down 