A few of these individuals are policymakers, who’ve been attempting onerous to answer the issues AI merchandise pose with out decreasing our capacity to harness their energy.
So firstly of this yr, my colleagues and I appeared world wide for indicators of how AI laws are prone to change this yr. We summarized what we discovered right here.
In China, one of many main strikes to be looking out for in 2024 is whether or not the nation will comply with within the European Union’s footsteps and announce its personal complete AI Act. In June of final yr, China’s high governing physique launched a listing of laws they have been engaged on. An “Synthetic Intelligence Legislation” appeared for the primary time.
The Chinese language authorities is already good at reacting to new applied sciences swiftly. China was in all probability the primary nation on this planet to introduce laws on generative AI mere months after ChatGPT’s massive break. However a brand new complete legislation may give China much more management over how AI disrupts (or doesn’t disrupt) the best way issues work at the moment.
However you shouldn’t simply take my phrase for it. I requested a number of consultants on Chinese language AI laws what they assume will occur in 2024. So on this e-newsletter, I’ll share the 4 primary issues they stated to anticipate this yr.
1. Don’t anticipate the Chinese language “AI Legislation” to be finalized quickly.
In contrast to earlier Chinese language laws that target subsets of AI equivalent to deepfakes, this new legislation is aimed on the complete image, and which means it’s going to take a variety of time to draft. Graham Webster, a analysis scholar on the Stanford College Heart for Worldwide Safety and Cooperation, guesses that it’s possible we’ll see a draft of the AI Legislation in 2024, “however it’s unlikely it is going to be finalized or efficient.”
One massive problem is that even simply judging what’s and isn’t AI may be so tough that attempting to sort out the whole lot with one legislation could also be impractical. “[It’s] at all times a query in legislation and tech whether or not a singular legislation is important, or whether or not it must be addressed by way of its functions in different areas,” says Jeremy Daum, who researches Chinese language legal guidelines on the Paul Tsai China Heart. “So a generative-AI content material regulation is smart, however simply AI? We’ll see what occurs.”
