EU AI Act takes effect August 2024, prohibiting harmful AI practices (e.g., subliminal manipulation, social scoring). Tips for compliance include focusing on prohibited use cases, early screening of new AI projects, and balancing governance with practicality.
TLDR: EU AI Act, effective August 2024, prohibits harmful AI practices: subliminal manipulation, social scoring, and biometric tracking in public. Compliance requires assessing current AI uses, screening future applications, and maintaining an efficient governance framework. Focus on high-risk, prohibited cases and leverage existing compliance processes to ensure adherence.
AI washing poses legal risks as companies exaggerate AI capabilities amid vague definitions in the EU's AI Act. Legal clarity is essential but increases scrutiny, with potential penalties for false claims. Companies face liability for misleading assertions to investors and customers. Developing accurate communications and verifying AI claims are crucial to mitigate legal repercussions.
https://www.clydeco.com/en/insights/2025/05/the-fine-print-of-ai-hype-the-legal-risks-of-ai-wa
Developers are braced for uncertainty as the EU AI Act is enacted, acknowledging that evolving compliance strategies will be necessary due to ambiguities in the regulatory landscape. Experts emphasize adapting governance frameworks while recognizing the gradual implementation of regulations. Companies with prior regulatory experience, particularly in data compliance, are better positioned to navigate the Act's requirements, while the finalization of a general-purpose AI Code of Practice remains delayed. The European Commission is exploring simplifications to the Act to ease compliance for businesses.
EU must create supportive AI regulations, address digital skills gaps, and ensure AI accessibility for all businesses to enhance competitiveness.
https://www.aboutamazon.eu/ai-and-competitiveness-shaping-europes-digital-future
European Commission analyzes stakeholder feedback on AI definitions and prohibited practices from public consultations, aiding in the application of the AI Act. Report highlights majority industry responses, calls for clearer definitions, and concerns over prohibited practices like emotion recognition and social scoring. Guidelines issued to assist stakeholders with compliance and will evolve based on feedback and new use cases.
1 in 3 IT leaders are reducing AI investments due to inadequate employee training and rushed implementations, leading to ineffective adoption strategies. Over half regret not training staff, while nearly 30% acknowledge hasty investments. Failure rates of AI projects are rising, prompting some firms to abandon initiatives. However, successful AI adopters report improvements in productivity and decision-making.
https://www.ciodive.com/news/IT-CIO-AI-regret-investment-experimentation-asana/747683/
EU's General Purpose AI Code release missed May 2 deadline; now expected by August, delaying related AI Act provisions. Reasons for delay include allowing feedback and assessing support from AI providers. Political solutions may be necessary if not finalized by August, while tech developers face additional regulatory challenges.
EU issues guidelines clarifying prohibited AI practices under AI Act. Key prohibitions include manipulative techniques, social scoring, risk assessments for crime prediction, untargeted facial image scraping, emotion recognition in certain settings, biometric categorization of sensitive traits, and real-time biometric identification for law enforcement. Guidelines establish legal certainty, refine definitions, and highlight the interplay with existing EU laws. Safeguards for exemptions will require impact assessments on fundamental rights.
States are enacting AI laws influenced by the EU AI Act. Common features include disclosure of AI-generated content, use-case transparency, regulations for high-risk applications, and anti-discrimination measures. States like California, Colorado, and Utah lead in these regulations, emphasizing transparency and stakeholder compliance, with potential sanctions for non-compliance. Companies must align with these laws through governance programs, risk assessments, and ethical practices.
https://www.corporatecomplianceinsights.com/states-passing-ai-laws-what-do-they-have-common/
