AI Compliance in Business and Technology

What We Covered
During the Webinar

During the session, we went through six documented AI incidents and what they cost the companies involved. We compared how the EU and the US regulate AI, mapped which obligations already apply, and walked through a 90-day governance plan that starts with one person and a spreadsheet.

Webinar agenda:

• AI risks: six documented incidents and what they cost the companies involved
• The regulatory landscape: the EU AI Act and GDPR vs. the three layers of US regulation
• Who the AI Act applies to: providers, deployers, and companies whose AI output reaches EU users
• The US picture: existing law applied to AI, state AI statutes, and the federal push to override them
• A 90-day AI governance plan: AI census, prohibited-use screen, production access, staff training, disclosure, and vendor contracts
• After day 90: impact assessments, human oversight, testing, logging, and incident response ahead of December 2027
• What compliance gives you: nine business benefits beyond avoiding fines

Key takeaways:

• Most companies run three to five times more AI systems than they estimate. The highest-risk one is usually a system nobody bought as AI: it arrived as a feature in a tool you already had.
• In the US, most enforcement runs through existing law. Discrimination, consumer protection, and contract rules already cover AI in hiring, credit, and customer service.
• If your chatbot makes a promise to a customer, your company is bound by it.
• You can close the AI Act’s banned-practices check and staff training by day 60, and the transparency duty (binding since August 2, 2026) by day 90.
• High-risk obligations take effect on December 2, 2027. From a standing start, that’s roughly 15 months of work.
• The controls that make you compliant are the same ones that protect your IP, your production data, and your money. Compliance comes with them.