Legendary founder of the Silk Road darknet marketplace Ross Ulbricht has published a tweet from prison. This time, he talked about the artificial intelligence technology.

Ulbricht has spent more than eleven years in prison now after being arrested in 2013 when Silk Road was shut down by law-enforcement agents.

Ulbricht praises AI tech

The Silk Road founder tweeted that he had read an article about experiments on AI software producing images on the basis of people’s thoughts. Ulbricht seemed to be quite enthusiastic about that development, stating that technology has become much more advanced since the times when he was still free.

It is likely that Ulbricht referred to the last year’s article published by Vice, whose title says: “Researchers Use AI to Generate Images Based on People's Brain Activity.” The article tells about how researches have been making experiments of making AI “reconstruct high-resolution and highly accurate images from brain activity by using the popular Stable Diffusion image generation model.”

The scientists claimed that they did not need to train or fine-tune the AI models to make them create those images.

Elon Musk calls for AI safety bill in California

Today, Elon Musk took to the X social media platform to make an important AI call.

Musk said he believes it is necessary that the California state should pass the SB 1047 AI safety bill. The xAI founder added that he has been an advocate for regulating AI in the past 20 years. Artificial intelligence development, Musk stressed, should be regulated by authorities just like they regulate “any product/technology that is a potential risk to the public.”

This bill authored by State Senator Scott Wiener would be the first bill that, if passed, would require AI creating companies to test their large-scale AI models for safety before releasing them. In response, large AI companies threatened to leave the California state should the bill be passed.

The document would force companies that are involved in creating large AI models worth more than $100 million to train, to “limit any significant risks in the system that they find through safety testing.”

AI developers will also be required to come up with a “technical plan” that would handle safety risks and “to maintain a copy of that plan as long as the model is available, plus five years.”