EU AI Act Amendments Finalized, Reshaping Data Governance for High-Risk Systems

European Union legislators have provisionally agreed upon significant amendments to the EU AI Act, with formal adoption by the European Parliament and Council expected by July 2026. This legislative update, often referred to as the “Digital Omnibus on AI,” introduces crucial clarifications and extended compliance deadlines for various AI applications, particularly those deemed high-risk. The agreement, reached on May 7, 2026, aims to streamline implementation and foster innovation while reinforcing robust data governance for AI systems across the bloc.

The amendments notably postpone compliance deadlines for high-risk AI systems, with obligations for standalone high-risk systems now due by December 2, 2027, and for high-risk AI embedded in products by August 2, 2028. This deferral provides additional time for companies to adapt to the complex regulatory landscape, particularly concerning testing, documentation, and third-party assessments for AI systems. Furthermore, the amendments extend simplified compliance frameworks for small- and medium-sized enterprises (SMEs) to include companies with up to 750 employees and €150 million in annual revenue. New prohibitions also target AI systems that generate non-consensual intimate content or child sexual abuse material, effective December 2, 2026.

Strategic Data Partnerships Drive AI Adoption

Beyond regulatory shifts, the data deals market continues to see significant activity, particularly in AI data partnerships. Thomson Reuters announced on June 3, 2026, that it is leveraging Snowflake's AI Data Cloud to power its enterprise AI and data platform, aiming to deliver trusted, governed intelligence at scale. This partnership builds on Thomson Reuters' existing use of Snowflake since 2021, having created a single, secure source of truth across over 37,500 governed tables and 350 data sources. Similarly, the Snowflake Summit 26, held in early June 2026, showcased a broad suite of AI product launches and expanded partnerships, emphasizing the company's pivot to position its Data Cloud as a central platform for enterprise AI.

Acquisitions and Funding Bolster AI Data Ecosystem

In the acquisition space, Databricks made headlines in March 2026 with the strategic acquisition of three AI security startups—Lakewatch, Antimatter, and Siftd—to enhance its AI governance and security capabilities. This move, part of a larger $5 billion funding deployment, addresses growing enterprise concerns over AI governance and security vulnerabilities.

Meanwhile, the data-for-AI funding landscape remains dynamic. Earendil Labs, an AI-driven biologics startup, secured $787 million in March 2026, accelerating its platform for antibody and biologic design. Similarly, Perplexity AI saw its valuation climb to $21.21 billion following its Series E-6 funding round in early 2026, highlighting continued investor confidence in advanced AI platforms.

Global Data Flow and Public Sector Initiatives

Efforts to facilitate secure and interoperable cross-border data flows are also advancing. Singapore's TradeTrust initiative, as highlighted in its June 5, 2026, newsletter, is actively promoting seamless interoperability for digital trade ecosystems, leveraging AI to enhance data processing and validation across supply chains.

On the public sector front, the UK Government continues to advance its National Data Library initiative, aiming to unlock and share public sector data for AI research and development. This ongoing effort, discussed in early 2026, includes launching new projects to test how public sector data can be better integrated to improve services.

Why it matters for data owners

The latest developments underscore a clear trend: data is the foundational asset in the AI economy. Regulatory frameworks like the updated EU AI Act emphasize the critical importance of robust data governance, transparency, and ethical use, particularly for high-risk AI systems. For data owners, this means an increased imperative to understand and comply with evolving regulations, but also new opportunities to monetize well-governed, high-quality datasets. Partnerships, acquisitions, and funding rounds continue to highlight the immense value placed on specialized datasets and AI-driven data solutions. Companies that can securely and efficiently manage, share, and leverage their data assets in compliance with global standards will be best positioned to drive innovation and capture significant value in this rapidly expanding market.