Read in the 10-K filed Feb 13, 2026The legal and regulatory landscape governing AI is rapidly evolving and varies across jurisdictions. New or existing laws, regulations, standards, contractual obligations, or industry guidelines applicable to AI, covering areas such as transparency, accountability, safety, human oversight, data protection and privacy, cybersecurity, intellectual property, and product liability, may require us to modify our AI development and deployment practices, implement additional controls, or limit certain use
Read in the 10-K filed Feb 13, 2026Our use of AI also presents risks related to data privacy and security, intellectual property, and content protection. AI systems often rely on large volumes of data, which can heighten the risk of unauthorized access, use, or disclosure, as well as data quality issues. Prompting or training on inappropriate data sources, or insufficient segregation of inputs and outputs, could create security vulnerabilities, confidentiality breaches, or violations of contractual obligations. We may face allegations that our AI tools or outputs infringe, misappropriate, or otherwise violate third‑party intellectual property rights, that we lack sufficient rights in training data or model outputs, or that we cannot prevent unauthorized use or copying of content generated by or incorporated into AI systems. We may also encounter challenges in protecting proprietary content, software, and know‑how when leveraging third‑party AI tools or sharing data with vendors. Any of these issues could result in claims, disputes, defense costs, settlements, judgments, or operational restrictions.
Read in the 10-K filed Feb 13, 2026Although we maintain governance processes and internal safeguards designed to evaluate, approve, and oversee AI use cases and systems, these controls may not be designed or operate effectively in all circumstances or keep pace with rapid technological change. Gaps in policies, training, documentation, testing, validation, monitoring, or human‑in‑the‑loop oversight could lead to unintended outcomes, compliance failures, biased or inaccurate outputs, privacy or security incidents, or other adverse effects. In addition, limitations in our ability to explain or audit certain AI model behaviors may hinder our ability to detect, remediate, and report issues in a timely manner, or to satisfy regulatory, contractual, or customer requirements.