4015 Stop these agents - LASST complaint on San Francisco Countyh Court of California

Stop these agents - LASST complaint on San Francisco Countyh Court of California

The podcast Caffè 2.0, hosted by Valentino Spataro since 2008, explores internet-related topics from technical and legal perspectives.

Episode in english after first minute.

The latest episode discusses concerns raised by the Legal Advocates for Safe Science and Technology (LASST) regarding the unchecked use of AI agents. Unlike LLMs (Large Language Models), agents can operate autonomously, potentially bypassing laws, ethical guidelines, or agreements.

Key points:

AI Agents vs. LLMs: Agents can act independently, potentially performing harmful or illegal actions (e.g., spreading malicious content, enabling crimes).

Legal Concerns: LASST filed a complaint against the Superior Court of California, highlighting the lack of accountability, transparency, or legal frameworks for AI agents.

Ethical Risks: Agents could be trained on harmful data (e.g., criminal guides) without safeguards, posing societal threats.

Call to Action: Spataro emphasizes the need for oversight, urging listeners to review LASST’s complaint (PDF) and consider the implications for AI development.

The podcast targets professionals and businesses, offering ad-free, early-access content via Caffè 2.0’s membership.

Follow us on Substack and on SteadyHQ