What 20 Marketing Leaders Actually Think About AI, Brand, and the Execution Gap
I hosted a private dinner in Dallas last night with some of the smartest tech marketing leaders in the region. Here's what they are thinking right now about AI, automation and the future of brands.
Last night I hosted the Dallas edition of the Tech Marketing Rewired dinner series at Haywire in Plano. No presentations or pitches. Just one Jeffersonian-style table, great conversation and food, and about 20 marketing leaders who weren’t afraid to share their concerns, questions and how they are “rewiring” themselves and their teams.
But first, a huge thanks to our sponsors at Sendoso for making these dinners possible. We couldn’t do this without you!
OK, we’ve started hosting these roundtables in both Austin and Dallas, and so far, at the end of the evening, I walk away with more than I came in with. Last night was no exception. The Dallas group did not hold back and did not disappoint. Plus, I made a bunch of new friends.
Here’s what we discussed and my take on each of the themes:
AI Amplifies What’s Already There, Good or Bad
This was probably the most repeated idea of the night, and it came up in a number of different ways.
The consensus: AI doesn’t fix a broken strategy. It scales it. If your positioning is weak, AI helps you produce weak content faster. If your messaging is vague, AI gives you a hundred vague variations in seconds.
One of the best moments of the night was a story about a job candidate who showed up to an interview with a beautiful AI-generated positioning deck and couldn’t explain a single decision behind it. Hadn’t talked to customers. Hadn’t done the research. Just fed the website into a model and called it strategy.
My take: The floor of marketing is rising fast because of AI. But the ceiling is still completely determined by the quality of the thinking behind it. The fundamentals, customer insight, unique positioning, understanding your buyer, aren’t going away. They’re becoming the differentiator. AI can help synthesize thinking and present new ideas, but so far, a human needs to be in control.
Marketing Teams Have Three Types of AI Users (And Only One Is Pulling the Weight)
One of the more data-backed observations of the night came from someone who has been tracking AI adoption closely across a services organization. Their breakdown:
~10% are AI Leaders who have pushed a specific function to the point where AI is running it for them
~30% are AI Designers who actively build and experiment with AI workflows
~60% are AI Followers who use it when asked and never on their own
The interesting part: the best AI adopters aren’t necessarily the most technical people on the team. The common trait across the high performers was impatience. They iterate fast, don’t care about being wrong, and keep pushing until it works. Speed over polish.
The idea that came out of this discussion: take your AI Leaders, let them push a workflow to full automation, then have them personally teach one-on-one with the followers. Not a webinar. Not a Loom video. Actual one-on-one. That’s when adoption actually moves.
My take: You probably have an AI Leader on your team right now who you haven’t fully deployed. Find them. Give them a mandate. Let them teach.
Also, I do think there is strangely also a need for patience when learning these tools… because when anyone uses them, it takes time and patience to learn how to best harness he power… but yes, once you figure it out, then the need for shedding manual work + impatience drives action.
The “Garbage In, Garbage Out” Problem Is Really a Training Problem
The room got into a great discussion about how companies are mandating AI usage without giving teams the skills to prompt well. Leaders are saying “use AI or fall behind,” which is right, but without showing people what good looks like, you just get a faster pipeline of mediocre output.
Someone made a great analogy: it’s like negative keywords in search. You have to train it. Tell it what you don’t want. Use the muscles you’ve built from years of marketing experience, just applied in a new place.
The companies doing this well are building prompt libraries, skills md files, and ways of working, then tying AI usage to performance metrics. Not just mandating it, but measuring the quality of outputs and making it part of how success is evaluated.
My take: “Use AI” is not a strategy. “Here’s what great AI-assisted work looks like on this team, and here’s how we measure it” is a strategy.
Also, these tools are changing fast, so training a team on prompt libraries has quickly been overtaken by taking SOPs and turning them into .md files. Now there’s Chat GPT 5.4 agents and Claude CoWork (not to mention Claude Code). So, it won’t be enough to build a training plan and let it work… You’ll need to update it fast and ongoing to make sure you and your team don’t fall behind.
The CMO Role Is Changing Shape, Fast
This one got dropped almost casually but was very surprising to me. At least one company in the room has eliminated the CMO function entirely and moved marketing to report into sales under a GTM structure. This wan’t a small company either.
It’s not an isolated data point. It seems we’re seeing it across the industry. Marketing and sales are being pushed into the same lane, with RevOps and GTM leadership increasingly owning the combined function.
The upside is tighter alignment and shared accountability. The challenge is that corporate marketing leaders used to operating with strategic distance are suddenly in a very different power dynamic.
My take: I don’t think the CMO title disappears. But the “CMO as brand steward” model is getting squeezed hard. The CMOs who survive this shift are the ones who can speak fluently in pipeline, revenue, and systems and not just campaigns and creative.
I also strongly recommend that all CMOs become EXPERT AI operators. Get into Claude CoWork or Claude Code. Build stuff. Make mistakes and push boundaries in your companies. You can’t lead unless you know how it all works.
Note - I am writing an entire series for CMOs called CMOs as Builders that walks through my Claude Code setup. Subscribe to it here.
Brand Is Now a Risk Management Strategy, Especially in SaaS
This was probably the most debated discussions of the night. I showed a video of my interview with Santosh Sharan that set this up. AI is collapsing the cost of building software. A category that used to have 15 vendors now has 150. If anyone can clone your product in two weeks, what does differentiation even mean?
Everyone seemed to agree that your brand is your moat. Not features. Not pricing. How your buyer feels about you before they ever talk to a salesperson.
Someone made a point I’ve been saying for years: the number one reason deals are lost at most tech companies isn’t price. It’s confidence. The technical buyer just didn’t feel confident? That’s an emotion. That’s brand.
The “turtle and the hare” framing came up. Brand is the turtle. Slow to build, impossible to measure quarter-to-quarter, but it always wins in the end. In a world where your product can be replicated in weeks, the turtle is the only long-term bet that holds.
My take: If you’re not actively investing in brand, you’re renting customers. Every new competitor that enters your category makes the decision to skip brand investment a little more expensive.
We also talked about the amazing rise of email provider, Resend (see below). Right now, we think of brand being something that a human feels. But what is missing from most brand planning is how to get the AI agent to prefer your brand. That is coming fast. The good news is a strong human brand + a strong AI agent brand (agents recommending your brand) is going to result in a lot of business success.
GEO Is the New SEO, and Most Companies Aren’t Ready
One of the most forward-looking conversations of the night was about how AI is changing discoverability. We talked a lot about GEO (Generative Engine Optimization) and how most are planning for optimizing their brands to be recommended by AI, not just ranked by Google.
A concrete example came up: a company called Resend, an ESP, has built their entire brand strategy around being recommended by AI coding assistants. Their website is essentially structured documentation, written for the agent, not the human. And it’s working. They’re getting recommended in Claude, ChatGPT, and Cursor constantly.
The stat that got everyone’s attention: one dinner guest mentioned that roughly 85% of AI brand recommendations come from earned media, third-party sources, Reddit threads, G2 reviews, industry publications, not your own website.
My take: PR is having a renaissance. If AI is pulling from earned sources to make recommendations, then every piece of press, every G2 review, every Reddit mention matters more than it ever did. Your earned media strategy is your GEO strategy.
Not far behind may be your agent PR strategy. Who’s getting our content ready for the AIs to “review” us all?
The Junior Marketer Pipeline Problem Is Real and Nobody Has a Clean Answer
The conversation toward the end of the night got a little philosophical around marketing roles and staffing. If AI is automating the entry-level work, how do junior marketers learn? The junior role has always been the apprenticeship layer of the industry. You do the tactical stuff, you learn from being wrong, and you eventually develop judgment.
If we automate that layer, we might be producing a generation of marketers who can run tools but can’t think. Nobody in the room had a clean answer. But everyone agreed it’s coming.
The most memorable moment was a story about a 14-year-old who created a genuinely absurd, hilarious, totally original animation. Two completely unrelated concepts stitched together in a way no AI would ever think to combine. The punchline: I challenge any AI to do that.
My take: Human creativity isn’t going away. But the path that teaches humans to be creative, doing the boring work, making the mistakes, developing taste, might be getting eroded. That’s worth paying attention to.
What’s Next for These Dinners
I’m committed to doing more of these. The format works. One long table, one conversation started by watching some amazing clips of tech marketing thought leaders from the podcast, no slides, no pitching. It’s really the antidote to the conference panel and/or the “networking dinner. So far, these dinners are really “rewiring” the conversation and the thinking after leaving the room.
We’re planning more dinners in Dallas and Austin throughout the year, and I’m actively looking at expanding to other markets. Top of my list right now: Chicago (strong B2B tech scene, underserved for this kind of format), New York (the density of marketing leaders is unmatched), theBay Area (a no brainer for this audience) and dinners around major events like SaaStr, Inbound, or Adobe Summit where the right people are already in one city.
If you want to be on the list for future dinners in any city, hit me up. These are always invite-only and space is always limited. That’s the point.
And if you were in the room last night in Dallas: thank you. You showed up, you shared your concerns and ideas, and you made it easy. That room had some of the best off-the-record insight I’ve heard in a long time. More to come.
Have thoughts on what topics should drive the next dinner? Want to bring this format to your city? Find me on LinkedIn or reach out to the team at Mighty & True.
If you need help closing the strategic gaps in your marketing organization, people, AI, channels, or execution, that’s exactly what we do at Mighty & True.