The FTC fines Cox Media Group and two partners $930,000 over an 'Active Listening' AI ad service that never listened

Verification status: CHECKING — handed to the checker, not yet graduated. NOT verified, NOT green. This is an AI-washing enforcement record, an honest-negative, not a client testimonial. No green badge is sought; the war-room never sets verified.

The problem

“Your phone is listening to you and selling what it hears” is one of the most persistent folk-beliefs about the ad-tech industry, and one marketing product turned it into a sales pitch. On 21 May 2026 the U.S. Federal Trade Commission announced it would require Cox Media Group and two smaller marketing firms to pay a total of $930,000 to settle allegations they deceived customers by “falsely claiming to offer an AI-powered service that could target localized ads based on conversations captured from consumers’ smart devices and that consumers had opted into such targeting” (source). In three separate complaints the FTC named Georgia-based CMG Media Corporation, doing business as Cox Media Group, and two firms it worked with, New Hampshire-based MindSift LLC and Wisconsin-based 1010 Digital Works LLC (source).

What was built

The companies branded the product “Active Listening” and claimed it “listened in on consumers’ conversations overheard by smart devices, in real time, to target advertising, according to the complaints” (source). The three companies said the service would let small businesses advertise to consumers in the businesses’ desired locations by using “a special algorithm to listen in on and detect pertinent conversations from smart devices” (source). An independent business newsroom, Quartz, described the same pitch, reporting that “the companies claimed it used a special algorithm to pick up conversations from devices and deliver precise ad targeting for small businesses” (source).

The outcome

According to the FTC, the service did none of what it claimed: “this service did not, in fact, listen in on consumers’ conversations or use voice data at all—nor did the service accurately place ads in customers’ desired locations. Instead, the service the companies provided consisted of reselling—at a significant markup—email lists obtained from other data brokers” (source). Quartz reported the same finding independently, that “the service did not listen to any voice data. Instead, customers got email lists bought from data brokers, sold back to them at higher prices, and the ads were not accurately targeted by location” (source). The FTC’s Bureau of Consumer Protection director Christopher Mufarrige said, “Not only did the product these companies marketed not do what they claimed it did, but they also misled potential customers by claiming consumers had opted into this service when it’s clear they did not” (source).

On the question of consent, the companies “claimed that consumers had ‘opted in’ by agreeing to the terms of service that people have to accept when downloading and using apps”; the FTC countered that “clicking through mandatory terms of service does not constitute ‘opt-in consent’ for such an invasive service or for use of consumers’ voice data from inside their homes,” adding that if the service had functioned as advertised, “this collection and use of consumers’ voice data without adequate consent would itself violate Section 5 of the FTC Act” (source).

Under the proposed orders, “CMG must pay $880,000 while both MindSift and 1010 Digital Works will each pay $25,000, which will be used to provide redress to CMG customers impacted by these practices” (source), figures Quartz reported as CMG paying “$880,000” and the two smaller firms each paying “$25,000, bringing the total settlement to $930,000” (source). The Commission “voted 2-0 to issue the proposed administrative complaints and to accept the consent agreements,” which were “subject to public comment for 30 days” before the FTC decides whether to make the orders final (source).

Weakest load-bearing source. The strongest and load-bearing evidence here is the Tier-1 primary: the FTC’s own press release announcing the three complaints and proposed consent orders, from which every dollar figure and every quotation above is taken verbatim. Its honest limit is that it is the agency’s summary of the action, not the signed consent orders or administrative complaints themselves, and it records the FTC’s allegations settled by consent, not a court’s finding of guilt (the respondents consented without admitting the allegations). The Tier-2 corroboration is Quartz, a general business newsroom rather than a legal-trade or primary docket source; it independently confirms the $930,000 total, the $880,000/$25,000/$25,000 split, the 2-0 vote and the “resold email lists” finding, but the consent-and-Section-5 language and the exact company details (states of incorporation, the “means and instrumentalities” second count) rest on the FTC primary, not on Quartz. The marketing line CMG’s site once carried (“It’s True. Your Devices Are Listening to You.”) is reported by Quartz citing Gizmodo and is kept here as context, not a critical claim.

How this was verified. Method: every figure is quoted verbatim from the U.S. Federal Trade Commission’s press release, “FTC to Require Cox Media Group, Two Other Firms to Pay Nearly $1 Million to Settle Charges They Deceived Customers About ‘Active Listening’ AI-Powered Marketing Service” (21 May 2026 — Tier 1, primary; the FTC is first party to the enforcement action). It was retrieved this session via its Wayback capture (web.archive.org/web/20260621050622/…) because ftc.gov refused automated fetches, and the cleaned text is saved to sources/ftc-press-release-2026-05-21.txt. Independent corroboration is Quartz (qz.com), “Cox Media Group is paying $880,000 to settle FTC charges over a fake AI ad service,” by Colleen Cabili, updated 26 May 2026 (Tier 2; Wayback 20260627031833; saved to sources/qz-cox-media-ftc-2026-05-26.txt), which reproduces the total, the split, the 2-0 vote and the reselling finding firsthand. No confirmation was sought from the respondents; the badge never depends on a subject confirming its own conduct. Verified 2026-08-25 (checking round).

Sources

1. U.S. Federal Trade Commission · FTC to Require Cox Media Group, Two Other Firms to Pay Nearly $1 Million to Settle Charges They Deceived Customers About “Active Listening” AI-Powered Marketing Service · 21 May 2026 · https://www.ftc.gov/news-events/news/press-releases/2026/05/ftc-require-cox-media-group-two-other-firms-pay-nearly-1-million-settle-charges-they-deceived — Tier 1 (primary; the enforcing agency’s own release announcing the three complaints and proposed consent orders; archived Wayback 20260621050622, cleaned text saved to sources/ftc-press-release-2026-05-21.txt).
2. Quartz (qz.com) · Cox Media Group is paying $880,000 to settle FTC charges over a fake AI ad service · Colleen Cabili · updated 26 May 2026 · https://qz.com/cox-media-group-ftc-settlement-fake-ai-ad-service-052626 — Tier 2 (independent business newsroom; corroborates the $930,000 total, the $880,000/$25,000/$25,000 split, the 2-0 vote, the Mufarrige quote and the “resold email lists from data brokers” finding; archived Wayback 20260627031833, saved to sources/qz-cox-media-ftc-2026-05-26.txt).

Related case files

• The SEC’s cease-and-desist order finding Presto Automation’s drive-thru “voice AI” was run by off-site humans — the same failure mode from the other side of an enforcement action: an “AI” product that did not do the AI thing it was sold as doing.
• The FTC’s action banning Air AI from marketing business opportunities over deceptive AI claims — the same regulator, the same year, again over an AI product oversold to small businesses.
• The SEC’s $310,000 AI-washing penalties against Rimar Capital — a parallel “AI-washing” enforcement measuring the gap between an AI marketing claim and reality, useful for pricing how regulators treat exaggerated AI capability claims.