To risk or not to risk? The "Brussels effect" on AI regulations in Latin America
summary
A rundown of concrete claims and surprises: a paper found Hollywood scripts in training data triggering strikes over copyright; EU impact work guessed 10% high‑risk uses then recent Europe research moves that to 18% while a Brazil study measures 32% of uses flagged high risk; Latin American bills often copy the EU risk‑based frame but add unique burdens like a Brazil‑only plug standard analogy, heavy compliance costs that push startups to sell, an open‑source exemption for some models, and alternatives like Japan’s copyright focus and Singapore sandboxes.
AI-written summary · about
chapters (21)
- 0:00AI regulation in Latin America
- 1:08Europe’s risk-based AI approach
- 4:21Brussels effect on global standards
- 6:06EU digital market strength
- 7:10EU AI Act’s risk framework
- 8:12EU digital legislation ecosystem
- 9:16Latin America adopts EU-style AI
- 10:43Challenges in AI regulation
- 11:50Anti-big-tech agenda
- 13:15Creating AI regulatory agencies
- 13:35Japan/Singapore AI approaches
- 14:40Global AI regulation balance
- 15:00AI regulation impact on innovation
- 16:06Open-source AI competition
- 17:09Regulatory burden on SMEs
- 18:35AI safety institutes
- 19:39Risks of incompatible standards
- 20:43Harmonized AI regulation needed
- 21:27Q&A: Brussels effect on economies
- 22:54Q&A: Sector-specific regulation
- 25:01Closing remarks
related talks
36mResponsible AI: In Theory, Practice, and Policies
25mThe Fine Balance of Regulation & Innovation
01h 10mGlobal Approaches to AI Regulation Towards an International Law on AI
26mThe European AI Act: Challenges and Opportunities
51mIntersection of International Relations and AI
01h 08mTHE GOVERNANCE OF AI: CONVERGENCE OR DIVERGENCE?
01h 06mRisking Fundamental Rights The Myriad Ways of Risk in EU Law and Policy
35mThe European AI Act: a data scientist’s perspective