The State of AI in Marketing: What's Real, What's Hype, and What It Means for Science & Tech

Everyone's Using It. Almost Nobody's Winning With It.

Adoption is done being a debate. McKinsey found generative AI use jumped from 33% of organizations in 2023 to 79% in 2025, with 88% now using AI somewhere in the business. Duke/Deloitte's CMO Survey says AI now touches roughly a quarter of all marketing work, nearly double the year before, and CMOs expect that to double again within three years.

Here's the catch: adoption isn't impact. McKinsey found only about 6% of companies are actually pulling measurable bottom-line value out of their AI spend. Gartner's 2026 CMO survey backs that up: budgets for AI are over 15% of marketing spend now, but only 30% of teams say they're actually mature enough to scale what they've built. Seventy percent want to be “AI leaders” this year. Seventy percent also admit their processes aren't ready for it. Same 70%, probably.

For science and tech companies, this gap tends to be even wider. Regulatory complexity and long sales cycles make “move fast and see what sticks” a much riskier strategy. That's also where the opportunity is: most competitors are stuck at “we bought the tools.”

The Real Shift Isn't Content. It's Who's Doing the Buying.

Early AI-marketing hype was all about writing faster. That's yesterday's news. Gartner projects that by 2028, 60% of brands will use agentic AI (systems that act on their own) for one-to-one personalization, and up to 90% of B2B buying could run through AI agents. McKinsey thinks agentic AI alone could drive over 60% of the $463 billion in productivity gains it sees sitting on the table.

More strikingly: your buyer may not be a human reading your website first anymore. Forrester found 89% of B2B buyers now lean on generative AI throughout their research process. G2 found 51% of B2B buyers start research inside an AI chatbot, up from 29% just a year ago.

That's the whole reason answer engine optimization (AEO) exists: making sure ChatGPT, Perplexity, and Google's AI Overviews actually know who you are and cite you correctly. Early data shows AI-referred traffic converts noticeably better than regular organic search, and content that gets refreshed regularly earns more AI citations over time. (Yes, that's also why this series is quarterly and not “evergreen.”)

If you're a deep-tech or biotech company whose buyer is a scientist or an investor doing diligence, this isn't a nice-to-have. If the AI summarizes your category and skips you, you've lost the shortlist before a human ever opens your site.

Where Science and Tech Actually Stand

The general numbers only tell half the story. Here's the sector-specific half.

NVIDIA's 2026 healthcare and life sciences report found 70% of organizations actively using AI (up from 63%), and 74% for pharma and biotech specifically. Generative AI/LLM use jumped from 54% to 69%.

But the adoption-impact gap? Even wider here. Deloitte found only 22% of life sciences leaders have actually scaled AI past the pilot stage, and just 9% report real returns from it, even though roughly 85% of pharma and biotech leaders increased AI spending in 2025. Spending was never the constraint. Execution was.

That tracks with what we see up close: the companies getting real value aren't the ones with the most AI subscriptions. They're the ones who did the unglamorous work (strategy, positioning, data foundation) that makes any marketing effort work, AI-powered or not.

The Takeaway

AI isn't the headline of your marketing strategy. It's the engine under the hood. It makes a good strategy sharper and faster. It can't manufacture a market position you don't have.

That's the lens we use internally too, and it's why we'll keep coming back with fresh data instead of writing one report and calling it done. Next up: how to actually measure whether AI search engines know your company exists, and what a starting baseline looks like if you haven't checked yet.

Frequently Asked Questions

What's the current adoption rate of AI in marketing?

88% of organizations use AI somewhere in the business (McKinsey), and generative AI adoption grew from 33% to 79% between 2023 and 2025. But only about 6% of companies report real, measurable business impact from it.

What is answer engine optimization (AEO)?

AEO is making sure AI tools like ChatGPT, Perplexity, and Google AI Overviews can find, understand, and correctly cite your company. It builds on SEO rather than replacing it, and matters more every year as buyers start research inside AI chat instead of Google.

Is AI adoption different in biotech and life sciences?

Adoption is actually strong: 70% active usage industry-wide, 74% in pharma and biotech specifically (NVIDIA, 2026). But the gap between adoption and results is wider here too: Deloitte found only 22% of life sciences leaders have scaled AI successfully, and just 9% see significant ROI, largely due to regulatory and data-quality hurdles.

How often should we revisit our AI marketing strategy?

Quarterly, at minimum. Especially for anything AEO-related, since research shows AI search engines favor content that's kept fresh.

Sources

Gartner (2026 CMO Spend Survey; Agentic AI research; Marketing Automation survey)  ·  McKinsey & Company (Global AI Survey; State of Marketing Europe 2026; 2026 Global B2B Pulse Survey)  ·  Forrester (B2B generative AI adoption research)  ·  Duke University / Deloitte CMO Survey (2026)  ·  G2 (The Answer Economy: 2026 B2B Buyer Behavior Report)  ·  NVIDIA (2026 State of AI in Healthcare and Life Sciences)  ·  Deloitte (2026 Life Sciences AI survey)