Which statement about AI externalities and policy tools is accurate?

• A. Positive externalities include productivity gains and knowledge spillovers; negative include privacy loss, bias, and displacement; policy tools include regulation, subsidies for safety research, antitrust action, and data governance.
• B. Positive externalities are harmful; negative externalities are beneficial; policy uses tariffs and licensing.
• C. Externalities only matter for corporations; policy uses tax breaks.
• D. Externalities are fixed and cannot be influenced by policy.

Answer: (A)

AI externalities are spillovers from AI use that affect people other than the one making the decision. These can be positive, like productivity gains and knowledge spillovers that raise overall welfare, or negative, such as privacy loss, bias, and displacement of workers. The statement that accurately captures this and ties it to policy tools is that policy aimed at AI should address both sides by using regulation to set safety and fairness standards, subsidies for safety and foundational research to encourage beneficial development, antitrust action to prevent market power and ensure competitive outcomes, and data governance to protect privacy and manage how data is used. This combination shows how policy can foster beneficial AI effects while constraining harms.

Other options mischaracterize the landscape: one reverses the signs (positive externalities as harmful) and suggests tariffs and licensing as primary tools, which isn’t aligned with how AI externalities are typically managed; another narrows focus to corporations and tax breaks, ignoring broader social impacts and a wider set of policy instruments; and another claims externalities are fixed and unchangeable by policy, which overlooks how incentives, regulation, and governance can shape outcomes.