Jim Sullivan’s book, “The Book on AI Doc Review,” argues that AI-powered document review can outperform humans in eDiscovery by improving classification accuracy. The author outlines methodologies for validating these reviews using metrics like recall and precision, emphasizing sampling and pre-validation to ensure quality results.
In 2026, organizations will adopt a disciplined approach to agentic AI, focusing on specific use cases, governance controls, and measurable performance outcomes to determine which projects progress from pilot to production.
AI models play a crucial role in decision-making for financial institutions, but their opacity presents compliance challenges. As reliance on third-party foundational models increases, institutions must navigate the complexities of transparency, validation, and regulatory requirements to maintain customer trust and operational integrity.
On December 19, New York became the first state to enact major AI safety legislation with the signing of the RAISE Act, following President Trump’s executive order on federal preemption of state AI laws. As federal enforcement narrows, states are imposing stricter compliance requirements, creating a complex regulatory landscape for companies deploying AI.
Elon Musk’s xAI has restricted its Grok AI system following backlash over misuse involving digital images of women and underage girls. OpenAI supports a California ballot measure regulating AI chatbot interactions with children, collaborating with Common Sense Media on a compromise. Additionally, new research highlights AI model memorization concerns, while Ford and the SEC announce AI-related initiatives.
The Justice Department is forming a new artificial intelligence taskforce to address inconsistent state AI regulations that may hinder AI company growth. This initiative follows President Trump’s executive order seeking a unified regulatory framework across the U.S.
Colorado’s groundbreaking Artificial Intelligence Act has been delayed to June 2026 as legislators reconsider its provisions amid concerns from the tech industry. This delay highlights the challenges of balancing consumer protection with innovation in pioneering AI legislation.
The National Institute of Standards and Technology (NIST) has released draft guidelines for applying its Cybersecurity Framework to organizations adopting artificial intelligence, inviting public comments by January 30, 2026. The Cyber AI Profile aims to help organizations manage cybersecurity risks related to AI systems and enhance their cybersecurity practices.
Aiomics has partnered with StreamedUp to launch a CME-certified curriculum focused on AI Literacy, as required by the EU AI Act. This course equips medical professionals with the skills needed to responsibly use AI tools in healthcare.
A federal representative has proposed reforming Mexico’s data protection law to regulate artificial intelligence (AI) technologies, aiming to establish mandatory rules for entities interacting with AI and address the current legal vacuum.