Comment from CtrlAltX Studios LLC

AI Policy Statement; Matter No. P264200 Docket No. FTC-2026-0859 COMMENT OF CTRLALTX STUDIOS LLC SUMMARY The proposed statement addresses state laws that pressure AI developers to alter accurate outputs. That concern is coherent as applied to laws imposing substantive duties regarding algorithmic outcomes. It does not translate to state laws that require only measurement and disclosure. We urge the Commission to distinguish these categories expressly, and to clarify that compliance with a disclosure or measurement mandate is not, without more, evidence of deception under Section 5. We offer New York City Local Law 144 as a worked example, with primary-source evidence from three years of operation. A fuller version is attached. INTEREST OF THE COMMENTER CtrlAltX Studios LLC is a New York technology studio. We are not an AI developer subject to the proposed statement, and we do not sell compliance services related to Local Law 144. Our interest follows from primary-source research we conducted into that statute's scope, compliance rates, and enforcement history. I. DISCLOSURE MANDATES AND OUTPUT-ALTERATION MANDATES ARE DIFFERENT INSTRUMENTS Local Law 144 imposes two obligations on covered employers: obtain an independent bias audit of an automated employment decision tool within the preceding year and publish a summary of the results; and notify candidates at least ten business days before using the tool. Neither requires anyone to alter an output. The statute mandates that disparate impact be measured, and that the measurement and the tool's use be disclosed. An employer may lawfully publish an audit showing significant disparities and continue using the tool. The statute compels information, not outcomes. This differs from a state law imposing a duty of care regarding algorithmic discrimination, which may create pressure to modify system behavior. The proposed statement's theory of harm - undisclosed steering contrary to consumer expectations - has purchase against the second category. It has none against the first, because a disclosure mandate produces disclosure, which is what the theory says consumers are owed. II. THREE YEARS OF OPERATION SUPPORT THIS READING Local Law 144 took effect January 1, 2023; enforcement began July 5, 2023. Compliance is minimal. A peer-reviewed study presented at the 2024 ACM Conference on Fairness, Accountability, and Transparency examined 391 covered employers and found 18 had posted a bias audit and 13 had posted a transparency notice. Enforcement has been effectively absent. The New York State Comptroller's audit of the enforcing agency, Report 2024-N-6 (December 2, 2025), found that the agency had reviewed 32 companies and identified one instance of non-compliance, while the auditors reviewed the same 32 companies and identified at least 17. The agency received two complaints across a two-year audit period, both concerning positions outside its jurisdiction. No civil penalty under the statute has been publicly issued. No output-steering pressure has materialized. We are aware of no evidence, and the proposed statement cites none, that Local Law 144 has caused any developer or employer to alter system outputs. That is unsurprising, because the statute does not ask them to. III. OVER-INCLUSION WOULD WORK AGAINST THE COMMISSION'S OBJECTIVE The proposed statement rests on the premise that consumers are entitled to know how an AI system has been steered. Disclosure mandates are among the few existing mechanisms that generate exactly that information. If the final statement can reasonably be read to place compliance with a state disclosure mandate under suspicion, it introduces uncertainty into obligations that are already poorly complied with. The likely effect is not less output steering - there is none here to reduce - but less disclosure. IV. RECOMMENDATION We respectfully recommend that the Commission add language to the following effect: "Nothing in this statement should be construed to suggest that compliance with a State law requiring measurement, auditing, or disclosure regarding an automated system - where that law does not require alteration of system outputs - constitutes a deceptive act or practice under Section 5. The concerns addressed here arise where system outputs are steered toward objectives not disclosed to, and not reasonably expected by, consumers." This preserves the statement's force against the conduct it targets while avoiding a chilling effect on transparency regimes. SOURCES: N.Y.C. Admin. Code sections 20-870 to 20-874; 6 RCNY sections 5-300 to 5-304 and the DCWP published Automated Employment Decision Tools FAQ; New York State Comptroller Report 2024-N-6 (December 2, 2025); Wright et al., "Null Compliance: NYC Local Law 144 and the Challenges of Algorithm Accountability," FAccT '24, DOI 10.1145/3630106.3658998. Respectfully submitted, Bryan Barrett Founder & Managing Director CtrlAltX Studios LLC