The New York Times filed its second lawsuit against an AI company—this time Perplexity for copyright infringement. Meanwhile, McDonald's had to yank an AI-generated Christmas ad after the internet collectively dry-heaved.

One's a legal beatdown that could reshape how AI companies operate.

The other's a branding faceplant that should make every CMO think twice before greenlighting their next "AI-powered" creative campaign.

The New York Times filed a copyright infringement lawsuit against AI search startup Perplexity on December 5th. This is the Times' second swing at an AI company—they're already mid-bout with OpenAI and Microsoft.

The accusation? Perplexity's AI-powered search engine scrapes NYT articles (including the paywalled stuff you're supposed to, you know, pay for), then delivers "verbatim or near-verbatim" reproductions to users. No permission. No payment. No shame, apparently.

Here's where it gets spicy: The Times claims Perplexity made over 175,000 attempts to access their website in August 2025 alone. After being blocked. After receiving cease-and-desist letters. After 18 months of being told to knock it off.

That's not a bug. That's a feature they're proud of.

Perplexity's head of communications, Jesse Dwyer, responded with what might be the most tone-deaf quote of the year:

Bold strategy, Sir. Let's see how it plays out.

Here's the thing about that logic: it's already not working. Back in September, Anthropic agreed to pay $1.5 billion to settle a lawsuit with authors who claimed the company trained its models on pirated books. That's the largest publicly reported copyright recovery in history. Billion. With a B.

The legal landscape isn't shifting—it's already shifted. And someone forgot to send Perplexity the memo.

Meanwhile, Perplexity's dance card is filling up fast. They're now being sued by the Chicago Tribune (filed the day before the NYT lawsuit—rough week), News Corp (Wall Street Journal, New York Post), Reddit (for "industrial-scale" scraping), Encyclopedia Britannica and Merriam-Webster (even the dictionary is mad at you), and Amazon (for unauthorized purchases through their Comet browser).

When Merriam-Webster is suing you, maybe it's time to look up the definition of "permission."

Even if you're not building AI products, this matters. The tools your teams rely on—for research, content generation, summarization—carry their own legal baggage. Perplexity's headlines today could become your procurement headache tomorrow. Smart organizations are monitoring the legal exposure of their AI vendors the same way they track any third-party risk. If you're not already asking "where does this tool source its data?"—start. And build the internal muscle to pivot quickly when the answer turns problematic.

ChatGPT is a superpower if you know how to use it correctly.

Discover how HubSpot's guide to AI can elevate both your productivity and creativity to get more things done.

Learn to automate tasks, enhance decision-making, and foster innovation with the power of AI.

Download the free guide

McDonald's Netherlands released a 45-second AI-generated Christmas commercial on December 6th. Titled "the most terrible time of the year," it depicted a parade of holiday disasters—presents flying off cars, people eating pavement on ice, cats destroying Christmas trees—before suggesting that McDonald's is the perfect escape from festive chaos.

Solid concept, honestly. Relatable. Everyone's had that Christmas.

The execution? Not so much.

Within days, the internet rendered its verdict: "AI slop."

Viewers called it "creepy," "soulless," and "deeply unfunny." Film critic Richard Roeper delivered the kill shot: "If they were going for creepy, depressing, deeply unfunny, clumsily shot, poorly edited, and inauthentic—nailed it!"

McDonald's pulled the ad within four days and issued a statement calling it "an important learning" as they explore "the effective use of AI."

Corporate for: We stepped in it.

But wait—it gets better. The production company behind the ad, The Sweetshop, released a statement defending their work:

The defense managed to highlight the exact thing people were upset about—displacement of human creative talent for an inferior result—while also confirming that AI isn't even saving that much effort. Seven weeks of sleepless nights for this?

McDonald's isn't alone in this particular dunk tank. Coca-Cola has been running AI-generated Christmas ads for two years straight—and getting roasted both times. Their 2024 "Holidays Are Coming" remake was called "soulless" and "devoid of any actual creativity."

Their 2025 response? Double down.

Coca-Cola's head of generative AI, Pratik Thakar, told The Hollywood Reporter: "The genie is out of the bottle, and you're not going to put it back in."

Sure. But consumers can still choose not to buy what the genie's selling.

This is a governance failure, plain and simple. Somewhere between "let's use AI for this" and "let's publish it," there should have been a checkpoint that asked: Does this actually meet our brand standards? Have we pressure-tested it with real humans? The lesson isn't "don't use AI for creative work"—it's that you need robust internal policies dictating when, how, and under what oversight AI gets used for customer-facing output. Without that governance layer, you're one "publish" button away from serving your customers soggy fries.

We're still deep in the AI hype cycle—but the rules of the game are changing. Accountability and governance are no longer optional add-ons. They're table stakes.

On the legal front, courts are ruling that training AI on copyrighted content without permission isn't automatically "fair use." The $1.5 billion Anthropic settlement was the warning shot. The NYT-Perplexity lawsuit is the follow-up. More are coming.

On the cultural front, consumers aren't impressed by AI for AI's sake anymore. The novelty wore off somewhere between the third uncanny valley Santa and the fifth "we're exploring AI" press release. People are judging AI-generated content by the same standards as everything else now—and increasingly finding it lacking.

Whether it's legal exposure from your AI vendors or reputational risk from AI-generated content, the answer is the same—robust internal governance. Know what tools you're using and where they source their data. Establish clear policies for AI in customer-facing work. Build review processes with actual humans in the loop. The companies that thrive won't be the ones moving fastest—they'll be the ones moving deliberately, with their eyes open.

Whether it's legal exposure from your AI vendors or reputational risk from AI-generated content, the answer is the same—robust internal governance.

Know what tools you're using and where they source their data. Establish clear policies for AI in customer-facing work. Build review processes with actual humans in the loop. The companies that thrive won't be the ones moving fastest—they'll be the ones moving deliberately, with their eyes open.

I've spent the past two years implementing AI policy and governance frameworks at the enterprise level. If your organization is navigating these questions—vendor risk, acceptable use policies, AI governance structures—I'm always happy to talk. Reach out anytime.

Keep reading and learning and, LEAD the AI Revolution 💪

Hashi & The Context Window Team!