Fifteen AI ads aired during Super Bowl LX. Combined spend: north of $120 million. Nearly half of all online mentions were negative, and consumer sentiment tells an even rougher story. The ads that worked told human stories. The ones that flopped let AI do the talking. If you're making decisions about how AI shows up in your business — your brand, your marketing, your customer experience — the Super Bowl just handed you a very expensive case study in what not to do.

Fifteen of the sixty-six Super Bowl ads were AI-related — either selling AI products or made with AI tools. That's roughly 23% of the entire broadcast.

At $8 to $10 million per 30-second slot, AI companies collectively spent more on Super Bowl airtime than any single tech category before them.

So how'd that work out? Let's just say the scoreboard wasn't very kind.

Google's Gemini ad, "New Home," took the top spot in Northwestern's Kellogg School annual advertising review — widely considered the gold standard for Super Bowl ad analysis. The ad showed a mom and her son using Gemini to imagine what their new house could look like. Different paint colors. A garden. A redecorated bedroom. Baby crawling around, family dog in the frame.

It was a product demo, technically. But it didn't feel like one — Google made it about a family going through a life change, with AI quietly helping in the background. The Kellogg panel gave it an A.

Anthropic's Claude campaign also performed well. Four spots that took direct shots at ad-supported AI — aimed squarely at OpenAI's move to bring ads into ChatGPT. One showed a guy asking a chatbot for workout advice and getting pitched shoes instead. Another had someone seeking relationship help and landing on a dating app recommendation. The tagline: "Ads are coming to AI. But not to Claude."

Anthropic saw an 11% jump in daily active users and a 6.5% bump in site visits — the biggest gains of any AI advertiser that week. Claude hit the top 10 free apps on the App Store.

What did Google and Anthropic have in common? Both ads were made by humans. Google leaned on classic emotional storytelling. Anthropic hired agency Mother to write genuinely funny creative. AI was the subject of the ads — not the thing making them.

Then there's Svedka.

The vodka brand ran what it called the first "primarily AI-generated" Super Bowl ad. Two robot characters dancing to a Rick James remix in a nightclub. Viewers called it "AI-generated nightmare fuel" and "Temu I, Robot." One commenter wrote: "I knew it was AI because a human with actual artistic ability wouldn't make something so ugly." Svedka disabled YouTube comments.

Adweek put it well: using AI to make the ad wasn't the idea. AI as the tool became the headline — and the result felt hollow rather than distinctive.

And then there's Ring, which turned into something much bigger than a bad ad.

Amazon aired a heartwarming spot about a lost puppy found through AI-enabled Ring cameras scanning the neighborhood. Cute kid, adorable Lab, happy ending. Except viewers immediately clocked the flip side: what happens when those same cameras scan for people instead of dogs?

Within days: the ACLU called it a wakeup call. Senator Ed Markey wrote an open letter to Amazon. The EFF warned about facial recognition. And Ring terminated its partnership with Flock Safety, a police surveillance company — a deal that had been in the works since October.

A 30-second commercial killed a business partnership in five days. That's not a bad ad. That's a governance failure.

My takeaway for you: the gap between how the industry thinks consumers feel about AI and how consumers actually feel is wide. At least for now. Don't take AI vendors' and consultants' advice on what you should be putting out there. Get a feel from your customers. Remember: AI vendors and consultants want to sell you stuff. But you have to keep your customers happy and keep your brand values alive.

One of our sponsors this week is giving you some practical advise on how to market properly with AI. Check them out.

61% of marketers say this is the biggest marketing shift in decades.

Get the data and trends shaping growth in 2026 with this groundbreaking state of marketing report.

Inside you’ll discover:

Download your 2026 state of marketing report today.

Get Your Report

This is a trust problem more than a marketing problem.

The companies that did better didn't hide the fact that they were selling AI products. Google showed Gemini front and center. Anthropic made AI the literal subject of its campaign. But both used human creativity to do the selling — real storytelling, real humor, real emotional hooks.

The companies that didn't? They led with "we made this with AI" as if the production method was the selling point. It wasn't. Consumers didn't care how the ad was made — they cared whether it was any good.

My opinion is that consumers aren't ready to be on the receiving end of fully AI-generated experiences. As much of an AI advocate as I am, I'm not ready for that either. The technology might be there, but the trust isn't. And you can't buy trust with a better model or a bigger ad budget. It gets built gradually, through experiences that actually deliver value to people.

What companies should be doing is a hybrid approach. Use AI to accelerate your internal workflows — drafts, data analysis, testing, ideation. Then let your people shape and deliver what customers actually see. AI behind the curtain. Humans out front. Not forever — but right now, while trust is still being established.

Brands that try to skip this step — the ones shoving AI-generated content in front of consumers who haven't been brought along for the ride — are going to keep getting Svedka'd. The ones building trust gradually, proving value quietly, and keeping a human in the loop where it matters? Those are the ones who'll still be standing when the dust settles.

This is what AI fluency looks like in practice: it's not just knowing how to use AI tools to create outputs. It's knowing when AI helps and when it hurts — with your customers, your brand, and your business. The leaders who understand that distinction and get ahead of it will build stronger trust with their audience and be in a much better position to grow.

I'm working on something new for The Context Window and I want to hear directly from you. If this newsletter has helped you in any way — understanding AI better, feeling more confident in a meeting, making a smarter decision, or just keeping up when the conversation turns to AI — hit reply and tell me in one sentence.

That's it. One sentence. I read every single reply.

P.S: Now don’t forget to hit reply and let me know your one sentence.

Keep reading and learning and, LEAD the AI Revolution 💪

Hashi & The Context Window Team!