According to reporting from the The Washington Post, at least 50 police officers are accused of using license-plate readers for tracking wives, girlfriends, exes and other forms of personal surveillance and nearly all of them involved Flock Safety's system. Flock, the $8.4 billion surveillance company, runs roughly 120,000 cameras around the country, reading license plates on behalf of law enforcement.

Nearly two weeks after the Post's initial reporting, Flock CEO Garrett Langley sat down for an interview with CBS News and said that he'd listened to an interview with one of the women who had been surveilled. "I apologize," he said. "It kills me that she went through that."

Later that same day, the company announced a series of solutions in a blog post including cutting the default data retention for new deployments from 30 days to seven, requiring case codes before every law enforcement search by the end of the year (with an override for emergencies like a missing child), and implementing an audit tool that flags unusual searches.

Langley then used the second half of the interview to move the argument onto ground he could win. "I don't think that Flock created police abuse," he said. "I think we're the first company to ever shine a light on it and build the tools to find it." So he's apologizing for what the cameras made possible while rejecting the premise that his company made it happen.

Now, there is fine print here. Existing agencies get to keep whatever retention period they already chose because the seven-day default is only binding for new customers. And the case codes and mandatory audit aren't live yet; both are planned to go live by the end of the year. By Flock's own count, a little over a third of its customers have the audit tool turned on today. So most of this package is promises and look, that's not nothing. A commitment with a date on it is something you can hold leadership to in a way "we take this seriously" never is, but it also means Flock just penciled in its own follow-up story. The ACLU's Chad Marlow said the retention change "may be a step in the right direction" but "whether this is a real change or just another Flock PR move, however, will depend on how its 'Evidence Mode' operates." Come December 31, people will be checking.

Equinox decided to ship two different statements to two different audiences within about a day of one another. The statement provided to CNN accused Robertson of manufacturing outrage "for clicks" and "intentionally ignoring" the campaign's full context. The next day, a fitness trade outlet got contrition, "we recognize there was an unintended impact and we apologize for that." By August 6, AsAmNews had run the two side-by-side and even the ad agency behind the campaign broke ranks, saying the critics were right and made a donation to an Asian American women's organization while the client itself was still swinging.

There's a piece of research that explains this gap, and it applies well here. Roy Lewicki, Beth Polin and Robert Lount tested the six components of an effective apology (regret, explanation, responsibility, repentance, repair, and asking to be forgiven) and found that people don't weigh each in the way we might assume. Regret ranked fourth of six in both of their experiments; "sorry" isn't even in the top three, and asking to be forgiven came dead last every time. What moves people is responsibility and repair: it's my fault, and here's how I prevent it from happening again.

Let's score this week's two statements against that list. Flock's regret was light and its responsibility was heavily hedged, since Langley said outright that his company didn't create the problem, but everything after that was repair including changing settings, deadlines, defaults, all the components a hostile reader can audit. Equinox on the other hand argued about whether there was anything to regret or even be responsible for.

A company doesn't get to decide that its critics' outrage is "manufactured." Let's remember Cracker Barrel, where outside researchers shared evidence that bots drove a large share of the logo backlash and they never once said so publicly. They ate it, reversed course, and Julie Masino was ousted as CEO. If the company holding the receipts can't play that card, Equinox never had it. Consumers and the audience hold that pen.

So before your next apology goes out, score your draft against those six components and weight your effort the way your key audiences actually do. One line of regret is plenty, don't ask anyone to forgive you, and spend the rest on repair work that's specific enough that even someone who distrusts you could check it out. The ACLU's question (about whether this is real change or a PR move) is the question every corporate apology gets asked, and repair is the only element that effectively answers it.

Many of us know that President Trump's teleprompter operator got caught this summer betting via Kalshi on the specific words Trump was going to say during a speech, which is a pretty darn reliable trade when you're the one physically loading the script into the prompter. But he wasn't the first. In a February advisory, the Commodity Futures Trading Commission (CFTC) walked through a case involving someone employed at Mr.Beast's YouTube channel who traded on script material he'd seen before it went out; Kalshi fined him $20k and suspended him for two years. These aren't example of financial masterminds, they're just people with early access to words.

If you open up Kalshi this morning markets are betting on what Walmart, Target, Nvidia, Dell and about a dozen other companies will say on their next earnings calls... I'm talking specific word. Ahead of DraftKings' call on August 7, you could put real money on whether its executives would say the word "Kalshi." That one traded near 12 percent, because of course nobody names their competitor out loud and nobody did. Later that day, CEO Jason Robins told Bloomberg TV what I suspect a lot of CEOs were thinking: "Making trades on whether someone will say something on an earnings call is probably not something that should be out there."

Law firms have written up their points of view already including Cleary Gottlieb in March, Sidley in June, Davis Polk in July and every memo centers around updating your insider trading policy so employees don't bet. Necessary, and also not sufficient. The earnings call is a comms product. We write the script, we build the Q&A doc, we run the rehearsal, we decide who gets the deck at eleven the night before. Everything the lawyers are worried about happens inside a vehicle we own, and I went looking for anyone in our field who has written about that and I'm not seeing much.

Last October, Brian Armstrong closed a Coinbase. earnings call by reciting "Bitcoin, Ethereum, blockchain, staking, and Web3," settling roughly $84,000 in open bets, then posted "lol this was fun." At the time it read as a CEO being a good sport but the CFTC is now examining whether these contracts are "readily susceptible to manipulation," and Kalshi has shelved its sports mention markets while that gets sorted.

But the earnings-call markets are untouched and still trading today. An earnings call is the moment a public company is legally required to tell all its investors the same thing at the same time; that's what the SEC's fair disclosure rule exists to protect. So if your CEO picks a word to settle a bet instead of informing the people listening, has the company misused that moment? Four law firms have written up employees betting on inside information but nobody has touched that one. I'd want my GC to have a view.

There's a tactical version of this that takes 15 minutes: have IR pull Kalshi's earnings-mention listings alongside the analyst previews, and if there's a live market on what your CEO is about to say, tell him before he goes on. Do that.

But the questions underneath it like who else has the script, and what that access is worth now doesn't get solved by a checklist, and it doesn't sit with the comms team alone. Our profession hasn't formed a view on this yet, and it will get formed for us if we wait, by a regulator, or by whichever company goes first and goes badly. So start forming yours where it counts, which is in front of your own CEO.