Faithful Catholic, U.S. Army vet, Bitcoiner. Daily Rosary warrior. Defending faith, family & freedom against the fiat dragons. Deus vult. Mary pray for us. โ๏ธ
Thousands of young Polish Catholics gather to enjoy electronic music.
The lyrics are:
โThe LORD is my shepherd; I shall not want. He makes me to lie down in green pastures: he leads me beside the still watersโ
Iโve had a number of conversations with folks inside and outside government about the current situation with Anthropic, and here is what I believe to be true:
โ As we know, Anthropic publicly released its Mythos class models earlier this week under the commercial name Fable.
โ Fable is Mythos with guardrails. But if those guardrails fail, then youโve exposed Mythos and its advanced cyber capabilities to people who shouldnโt have them. (Keep in mind that Anthropic itself widely promoted the idea that Mythos was a cyberweapon and needed to be regulated as such. They asked for government regulation of Mythos and championed the guardrails on Fable. If there is a vulnerability โ big or small โ it is Anthropicโs responsibility to patch.)
โ A highly credible trusted partner of both Anthropic and the USG who was testing Fable came forward with a jailbreak of those guardrails. The Admin asked Dario to fix the jailbreak or de-deploy the model. Dario refused.
โ In their blog post, Anthropic defended its decision by saying the jailbreak isnโt serious. That is not what the trusted partner and the USG believe; nor is that kind of minimizing language consistent with Anthropicโs brand as the AI safety company. Itโs difficult to fathom how they could claim a jailbreak allowing operability of a cyber weapon could be defined as not โserious.โ
โ In the past, Anthropic has always said that safety must be top priority and taken super seriously. In this case, Anthropic prioritized the continued offering of the consumer model over safety.
โ In reaction, the Admin issued the export control. The Admin did this reluctantly. Itโs been very surprised that Anthropic hasnโt wanted to cooperate with a reasonable safety request (ie fixing the jailbreak issue). Anthropicโs reaction is very much at odds with their branding and ethos as a safe AI research community.
โ The Adminโs hope now is that Anthropic remediates the safety issue, the export control is lifted, and Fable goes back into general release. The Admin wants all of this to happen as soon as possible. It is frankly bewildered that Anthropic hasnโt wanted to comply with safety requests that it previously said were its highest priority.
โ Those trying to misdirect and tie this action to the prior DoW/Anthropic issues are wrong. The Admin values Anthropicโs technical capabilities and feels that this issue, while serious, should be easily resolved. The ball is in Anthropicโs court.