Illinois SB 316 would create an unworkable regulatory regime through a sweeping definition of “AI companion” that captures enterprise tools and business software with no connection to the bill’s stated concerns, a constructive knowledge standard that perversely incentivizes more invasive data collection and a product liability preservation clause that exposes companies to unlimited liability for unpredictable AI behaviors that no developer can reasonably foresee. The bill’s decentralized enforcement structure — spread across State’s Attorneys statewide plus a private right of action — compounds the problem by making consistent compliance effectively impossible.
NetChoice Testimony in Opposition to Illinois SB 316, the Artificial Intelligence Companion Model Safety Act
May 27, 2026
Dear Chair Williams, Vice-Chair Rita and Members of the House Executive Committee:
NetChoice submits this testimony in respectful opposition to Senate Bill 316. While we strongly support Illinois’s interest in responsible artificial intelligence governance and the protection of young users from AI-related harms, SB 316 as currently drafted would create an untenable regulatory and liability regime that will stifle AI innovation, harm consumer access to beneficial technologies and ultimately disadvantage Illinois residents relative to consumers in other states.
NetChoice’s member companies are among the world’s leading technology innovators. Our members are deeply committed to developing safe and beneficial AI systems. Rather than building upon these existing private sector solutions, SB 316 pursues safety through mechanisms that are fundamentally flawed. The bill imposes an overbroad scope that captures services with no connection to child safety. It creates perverse incentives for invasive data collection. Most critically, it layers unlimited product liability on top of statutory requirements in ways that create unmanageable liability exposure for any company operating in Illinois.
SB 316’S SCOPE OVERREACH CAPTURES SERVICES UNRELATED TO CHILD SAFETY
SB 316 defines “artificial intelligence companion” so broadly that it will impose compliance obligations on services with no meaningful connection to the bill’s stated safety concerns. The definition applies to systems designed to simulate ongoing human-like relationships through retained user information and personalized interactions. This sweeps in business-to-business services, enterprise communication tools, workplace collaboration platforms, customer relationship management software, productivity applications, cloud infrastructure services and numerous other professional and operational technologies that have nothing to do with social or emotional harm to minors.
Under this overbroad definition, a productivity application that learns a user’s work preferences becomes an “AI companion” subject to requirements designed for social chatbots. Enterprise software that retains user information for operational purposes triggers the same obligations as consumer-facing emotional support systems. A business intelligence tool that generates personalized reports could be classified as an “AI companion.” These services pose no meaningful risk to children, yet they would be subjected to expensive compliance burdens that serve no safety purpose.
SB 316’S KNOWLEDGE STANDARD MANDATES PRIVACY VIOLATIONS IN THE NAME OF PRIVACY PROTECTION
SB 316 establishes a “constructive knowledge” standard that creates a fundamental contradiction: companies must collect more personal data to comply with privacy requirements. The bill deems a company to know a user is a minor based on any information, marketing inference or age assignment the company makes from any source. This creates a perverse incentive that directly contradicts consumer privacy interests.
To minimize their liability exposure, companies will rationally choose to collect, retain and analyze extensive personal information, including age verification data and detailed user identification information. The bill, therefore, mandates more invasive data collection practices in the name of protecting privacy. This is unworkable and will harm the very consumers the bill purports to protect. Companies that invest resources in privacy-protective approaches—such as limiting data collection—will face the same liability exposure as those that do nothing, removing any incentive for privacy-protective innovation.
SECTION 30’S PRODUCT LIABILITY PRESERVATION IS AMONG THE MOST PROBLEMATIC PROVISIONS IN THE BILL
Perhaps the most concerning aspect of SB 316 is Section 30, which explicitly preserves existing product liability causes of action, stating that “the remedies provided in this Act do not preclude any other lawful civil, administrative, or criminal remedy available under state or federal law, including product liability actions.” This provision is unworkable and will make it economically impossible to develop and deploy AI systems in Illinois.
Product liability law imposes liability for design defects, manufacturing defects and failure to warn—standards developed for physical goods with finite design variables and predictable failure modes. AI systems operate through complex internal algorithms and “emergent properties” that arise from interactions the creators did not specifically program. No company can foresee all possible outputs from complex machine learning systems. Subjecting AI to product liability standards means companies could be held liable for harms caused by properties that emerge unpredictably from the interaction of millions of parameters and novel user inputs.
This standard inverts basic principles of tort law, which ordinarily protect parties who take reasonable precautions and act with due care. Under SB 316, a company that implements safety features, fixes security vulnerabilities or improves its AI systems faces unlimited liability for completely unforeseeable consequences of those improvements. This creates a perverse incentive: companies that invest in safety, alignment and ethical considerations are penalized with no liability reduction, while companies that do nothing face the same liability exposure. This discourages safety investment and innovation.
By preserving product liability claims while simultaneously creating comprehensive statutory remedies and multiple enforcement pathways, SB 316 invites duplicative and conflicting litigation that makes compliance impossible. Plaintiffs can pursue claims under the statutory framework while simultaneously alleging product liability, leading to redundant and inconsistent standards. Companies cannot reliably forecast their legal obligations because conduct that satisfies the statutory requirements could still expose them to product liability claims based on unrelated or unforeseeable AI behaviors. This unlimited liability exposure for unpredictable AI system properties will chill investment in safety features and discourage responsible AI development.
SB 316’S DECENTRALIZED ENFORCEMENT STRUCTURE CREATES REGULATORY CHAOS
Beyond the product liability problem, SB 316 creates an extraordinarily complex and decentralized enforcement structure that will produce inconsistency, duplicative investigations and regulatory unpredictability. The bill incorporates violations into the Illinois Consumer Fraud and Deceptive Business Practices Act, grants enforcement authority to State’s Attorneys throughout the state and creates a private right of action for actual damages.
Without centralized authority, different State’s Attorneys will interpret and enforce the bill’s requirements inconsistently, leaving companies unable to understand what conduct satisfies their legal obligations. Companies may simultaneously face investigations from multiple State’s Attorneys for the same conduct, with potentially contradictory findings and remedies. Private litigation over technically complex and rapidly evolving AI systems will create an unpredictable legal environment where companies cannot reliably forecast their obligations. The complexity of AI technology combined with the incentives of private litigation will lead to inconsistent interpretations and unpredictable liability exposure. This fundamental uncertainty discourages development and deployment of innovative AI products and services.
The combination of decentralized government enforcement and private litigation creates a moving target that prevents companies from achieving stable compliance through good-faith efforts. The result will be either immediate non-compliance across the industry due to practical impossibility or withdrawal of AI services from Illinois to avoid liability exposure. Neither outcome serves Illinois consumers.
CONCLUSION
SB 316 is well-intentioned but fundamentally flawed. While NetChoice strongly supports the goal of protecting consumers, particularly young people, from AI-related harms, SB 316 will not advance that goal.
NetChoice urges the Senate to reject SB 316 and instead work with our members and the broader technology community to develop targeted and effective AI safety legislation. Responsible technology companies support reasonable AI governance. But effective regulation requires clarity, achievable standards and realistic implementation timelines. SB 316, as currently drafted, fails on all three counts. As always, we offer ourselves as a resource to discuss any of these issues with you in further detail, and we appreciate the opportunity to provide you with our thoughts on this important matter (The views of NetChoice expressed here do not necessarily represent the views of all NetChoice members.).
Sincerely,
Amy Bos
Vice President of Government Affairs, NetChoice
NetChoice is a trade association that works to make the internet safe for free enterprise and free expression.