- Bear and the Bull
- Posts
- AI is creating ads you cannot skip
AI is creating ads you cannot skip
Native ads are the new storylines - and AI is writing the script

Sally sells more leftward curving seashells by the seashore than rightward curving ones.
I have to start this week’s newsletter with “thank you.”
So many of you who shared your positive reaction to my write-up last week on the women’s digital health opportunity being more than “womb tech.”
You messaged me in English and in German, you shared with your friends and family, and I couldn’t be more pleased. I write about the things that seem so obvious that their lack of recognition in spheres of legitimate discourse feels brazen. Sometimes that feeling can make you question your sense of truth. So thank you for filling the once empty spaces with dialogue.
This weekend I went to the beach, and started assembling the smalls seashells I encountered near me.
I noticed something curious.
I found almost twice as many seashells that had curved to the left than those that curved to the right.
Is this statistically significant? My sample size of n < 100 would certainly point to “no.” Regardless, it was hard to ignore.
That is kind of how this week’s newsletter came together.
While watching some of my favorite shows, I started picking up on a different pattern — a wave of native ads, subtly woven in between the scenes. Once I saw one, I started seeing them everywhere.
So this week, we are diving into the rise of AI in ad tech - what works, what does not, and where I see real innovation happening.
📺 All ads are created equal - but some are more equal than others.

What this season of “The Bear” lacked in quality it made up for in quantity of native ads.
This weekend I started season 4 of The Bear.
The writing did not hold my attention. But the ads? They did.
From cookware cameos to culinary partnerships, the native advertising in this season was relentless — and, frankly, impressive.
The Bear is not alone.
Think vesper martinis in White Lotus, BMW’s placement in Hacks, or Corona in The Last of Us. These ads blend into the story. They do not interrupt the content; now, they are the content.

The vesper martini was a mainstay of White Lotus’ season 3 advertising.
One thing you might not know about me is that I love studying advertising. At Kellogg, one of my favorite courses was Advertising Strategy. I could gladly spend hours analyzing advertising copy, tone, placement, frequency, and the myriad other characteristics that make us feel something we can see an ad that just “works.”
In my job as an investor, I’ve been somewhat surprised to see more and more startups purporting to incorporate AI to automate advertising. Some even go as far as attempting to replace full-fledged creative teams with AI.

Others approach use a more streamlined approach. One company called Sotto is attempting to improve the SMS channel. The founding team previously built and exited another successful company in the media space, and have reunited to revolutionize how brands engage with customers and prospects on their phones.
Notably, this isn’t my first go-round in this space. Two years ago I wrote about a company called Memorable, which harnessed the power of gen AI to allow marketing teams to optimize marketing spend.
Last year, that company was acquired by Reddit.
So maybe I know a little bit of what I’m talking about,
Which brings me back to the bigger question: what kind of advertising do we actually want? This moment in AI advertising feels a lot like early programmatic: full of potential, but still missing the nuance.
Here are 5 shifts I am watching in AI-driven ad tech:
1️⃣ Synthetic creative generation with brand safety guardrails
AI can produce endless iterations of ad copy and visuals. I believe that the winners will be the tools that ensure alignment with brand tone, context, and compliance.
2️⃣ Dynamic creative optimization that actually adapts
Not just A/B testing headlines, but live responsiveness to user behavior, sentiment, and channel context — think: TikTok vs. LinkedIn, in real time.
3️⃣ Privacy-forward targeting without cookies
The best AI tools will unlock performance gains even as third-party data disappears. Look for startups building on zero-party and contextual signals.
4️⃣ SMS and owned channel optimization
As CAC (customer acquisition costs) rises, brands are re-centering on owned channels like SMS and email. AI that can adapt messaging cadence, tone, and format for each user will be a core advantage.
5️⃣ AI-powered measurement that ties brand to performance
We have long treated brand and performance as opposites. The next frontier will quantify the brand halo and ROAS in the same dashboard — with predictive lift models built in.
Your turn ⬇
What’s your favorite example of a native ad? Share in the comments below.
🎙 Content Recap
🎧I wasn’t able to release a new episode of Money Memories last week since I was on the road in San Francisco. New ones will be released this Wednesday, but until then, there’s almost 100 episodes for you to pick and choose a favorite for
📍 Where I’ll Be / Where I Want to Be
If you are attending any of these upcoming events, let me know. I would love to find time to connect:
August TBD: I’m cooking up an exciting women’s digital health collaboration with AllThingsFemtech. More to come!
29 September - 2 October: I’ve registered for Sibos’ Frankfurt conference. If you’re attending, drop me a note.
If you’re in any of these places, add these to your radar:
🔗 Other Interesting Reads & Listens
📌The Woman Who Spent 500 Days in a Cave: The New Yorker is the gold standard of long form journalism for me.
📌Wimbledon Adjusts to Live Without Line Judges: To know me is to understand that my love of tennis runs deep. It’s been really fascinating to see how the sport has incorporated advances in technology. Next step for Wimbledon: abolishing the curfew? I’m not sure myself, I do love tradition.
Till next week,
🎾 Ilona
Reply