Privacy, copyright, safety, governance, and regulatory developments.
- What happened: Canadian employers are increasingly using AI to monitor workplace safety and risks. - Why it matters: These tools can improve safety but may also raise privacy and mental health concerns. - What to watch: Experts advise HR leaders to involve employees in the implementation of surveillance tools.
- What happened: Future teachers shared their views on the ethical use of AI tools in education. - Why it matters: Understanding these perspectives helps shape guidelines for AI integration in teaching practices.
House lawmakers are negotiating federal legislation to block certain state laws on artificial intelligence, particularly those in California and New York that require developers to disclose information about new models. The proposed bill would allow states to regain regulatory power after two years, while discussions continue over whether to implement mandatory vetting for AI developers.
A House of Commons committee amended Bill C-16 to broaden the offence targeting non-consensual sexual deepfakes, including nearly nude images and explicit references to AI software. The change signals how Canadian law is adapting to generative AI harms and raises compliance expectations for platforms and tool providers.
Greece is updating its constitution to ensure that artificial intelligence serves human society. Prime Minister Kyriakos Mitsotakis described these changes as essential for protecting future generations.
The Cloud Security Alliance expanded its work on governance and assurance for agentic AI systems, aiming to make controls more practical as agents move into production environments. For enterprises, this is another sign that “agent governance” is becoming a recognizable compliance and risk-management category, not just an internal best practice.
CAISI signed frontier AI testing agreements with Google DeepMind, Microsoft, and xAI. This continues the shift toward government access to unreleased models for safety and national-security evaluation.
US officials will evaluate models from Google, Microsoft, and xAI before public release, adding a more formal pre-release review step for frontier AI. The move is important because it shifts some safety assessment upstream, before models reach broad public use.
Okta researchers report that agentic systems can be manipulated through prompt injection and “agent-in-the-middle” patterns that lead to credential exposure, even when organizations believe controls are in place. The takeaway for enterprises is straightforward: AI agents need the same kind of security architecture as other privileged software, not just better prompts.
New polling cited by the New York Times shows broad bipartisan concern about AI across jobs, energy prices, education, privacy, and mental health, with even younger voters' enthusiasm cooling. The findings suggest AI anxiety is moving from elite discourse into mass political salience and may shape the legislative agenda heading into the next election cycle.
The Connecticut House passed a comprehensive AI bill that Governor Ned Lamont is expected to sign, making the state one of the first to enact broad AI oversight. The move bucks active federal pressure from the Trump administration to leave AI regulation to Washington and signals continued state-level leadership in the absence of federal AI legislation.
Negotiations surrounding the European Union's AI Act Omnibus collapsed on April 28, raising concerns about upcoming compliance deadlines for high-risk systems. Key issues include potential exemptions for AI systems in regulated products.