Key takeaways
AI adoption is not AI replacement: AI agents take over execution such as content variants, targeting and testing, while people own direction, approvals and strategy.
According to McKinsey's State of AI 2025, 88 percent of organizations use AI in at least one function, yet nearly two thirds have not started scaling it across the enterprise [1][2].
Gartner expects over 40 percent of agentic AI projects to be canceled by the end of 2027 [3]. A lack of direction costs more than a lack of technology.
When intelligence becomes a commodity, vision, ambition and real consumer knowledge become the scarce resource in marketing.
Employees come along when companies communicate transparently, enable people before rolling out tools and start with small, measurable pilots.
Fear or momentum: will AI replace marketers?
Will AI replace marketers? No. AI replaces tasks, not the people who decide what a brand stands for and who it wants to move. Work is changing and tasks are shifting. Companies that roll out AI purely as a cost-cutting program lose exactly the people they need to carry the change.
Two moods are colliding in many German companies right now. The first is fear: employees regularly read headlines about job cuts justified by AI. A company survey reported by HR Dive found that nearly 4 in 10 companies plan to replace workers with AI by 2026 [6].
The second mood is momentum. For many marketers, this is the most exciting stretch of their careers. Ideas that used to lack budget and weeks of lead time can now be tested in an afternoon. Both groups use the same tools. What separates them is mindset.
Volto AI takes a clear position: AI adoption is not AI replacement. AI adoption is the integration of AI into workflows so that people can do more and better work with it. AI replacement is the substitution of human work with AI, primarily to cut costs. This guide shows what the data says, which tasks are shifting and how to lead your team through the change.
How far has AI adoption in companies really come?
Almost every company uses AI, but only a minority scales it with measurable impact. The gap rarely comes down to technology; it comes down to a lack of direction.
McKinsey's “The State of AI” 2025 is based on a survey of nearly 2,000 organizations [1]. 88 percent use AI in at least one business function, up from 78 percent the year before [2]. Yet nearly two thirds of respondents have not started scaling AI across the enterprise [1]. 62 percent are at least experimenting with AI agents [1].
Two more findings point to the same gap between adoption and impact. On June 25, 2025, Gartner predicted that over 40 percent of agentic AI projects will be canceled by the end of 2027, partly due to escalating costs and unclear business value [3]. According to Fortune, MIT's “The GenAI Divide” report concluded in August 2025 that 95 percent of generative AI pilots at companies deliver no measurable return [4]. The MIT study's methodology is contested, so treat the number as a directional signal rather than a precise value.
And in Germany?
According to Bitkom, 57 percent of companies in Germany use AI in 2026, a majority for the first time, up from 36 percent a year earlier and 20 percent in 2024. Yet 59 percent say they are not tapping its potential at all, and only 3 percent say they use it rather strongly (survey of 603 companies with 20+ employees, July/August 2026) [9]. Adoption is here; impact is not yet.
At the same time, concern is rising. In the KI-Monitor 2026 by TU Darmstadt and YouGov (more than 2,000 respondents), around 43 percent of people with very good AI knowledge expect AI could soon take over their work, and concern has risen across all occupational groups since 2025 [10]. The headlines add fuel: DeepL, of all companies a German AI success story, announced in May 2026 that it would cut around 250 jobs, about a quarter of its workforce, citing AI [11].
One detail from the same Bitkom study stands out: companies that already use AI are less likely to expect significant job cuts (9 percent versus 15 percent among non-users) and twice as likely to expect headcount growth (18 versus 9 percent) [9]. Those who use AI tend to see growth, not replacement.
Volto AI draws a sober conclusion from this: German companies don't have a tool problem. They have a direction problem. Rolling out tools without deciding which decisions should get better leads straight to pilot purgatory.
Why is AI adoption not AI replacement?
AI adoption is not AI replacement because AI agents take over tasks, not accountability. Execution gets cheap; direction stays scarce.
Agentic marketing is marketing in which AI agents independently carry out tasks such as content creation, campaign delivery and testing, while people set goals, brand guardrails and approvals. The day-to-day of a marketing team shifts as a result: less production, more steering and judgment.
The labor market paints a nuanced picture. The World Economic Forum's Future of Jobs Report 2025 expects a net 78 million additional job opportunities worldwide by 2030 [5]. At the same time, some activities disappear and entry-level roles come under pressure: according to the Stanford Digital Economy Lab, employment of 22- to 25-year-olds in AI-exposed occupations in the US now stands 19 percent below that of less-exposed peers, mainly because fewer are being hired [12]. Both sides are real. Leaders who only tell their team one of them lose credibility.
The table below shows how marketing tasks are shifting:
Area | What AI agents take over | What people newly own |
|---|---|---|
Content | Producing copy, image and ad variants | Defining the brand voice, setting quality thresholds, giving final approval |
Targeting | Delivering to audiences, shifting budgets in real time | Deciding which segments and motives the brand addresses |
Testing | Setting up and analyzing A/B tests, scaling winners | Framing hypotheses, translating results into strategy |
Reporting | Generating dashboards and summaries | Deciding which metrics matter and what follows from them |
Research | Compiling desk research and competitor overviews | Gathering and interpreting real consumer voices |
Strategy | Modeling options and scenarios | Setting vision and positioning, weighing risks |
The right-hand column can't be automated because it requires judgment: about the brand, about risk and about people. Volto AI focuses on exactly this column, because good judgment depends on good consumer knowledge.
Why are vision and ambition becoming the scarce resource?
When intelligence becomes a commodity, judgment decides: a clear idea of what you want to build. Alexandr Wang put it this way at Y Combinator: “The scarce resource isn't going to be intelligence or agency. It's going to be vision and ambition.” [13]
Infrastructure waves follow a recurring pattern: the scarce resource moves one layer up. When compute was scarce, GPU access decided who won. When model quality was scarce, research talent decided. When intelligence is abundant, the winners are those who know what to use it for.
Electrification is the historical precedent. When power became abundant, the utilities didn't capture the prize. The winners were the ones who knew what to plug in.
[BILD 2 hier einfügen: vision-knappe-ressource-steckdosen.webp – Alt-Text: A wall of grey sockets where only one person plugs in a colourful invention while the others stand unsure.]
Applied to marketing, that means AI agents can execute, but direction, brand understanding and real knowledge of consumers come from people. An agent without that knowledge optimizes for the average. It produces what sounds plausible, not what works with your audience.
Example: why plausible isn't enough
Volto's Switching Behavior 2026 study makes this concrete. In January 2026, Volto AI conducted 1,000 AI-assisted voice interviews with energy customers in Germany, weighted to be representative. 67 percent name a concrete breach of trust as a sure trigger for switching, not price [7]. 61 percent reject actively hunting for discounts and want fair, transparent pricing [7].
An agent relying on generic best practices would still push discount campaigns in this market. Volto's Dynamic Tariffs 2026 study (1,000 AI-assisted voice interviews, Germany, February 2026, representatively weighted) offers a second example: 70 percent worry about extra costs rather than missed savings, and even well-informed respondents share that worry at 69 percent [8]. A campaign promising “Save up to …” therefore targets the wrong motive.
Judgment comes from real consumer voices, not from plausibility. Volto explains why synthetic personas can't close this gap in Synthetic Personas: Plausible, but Not Representative.
What does this mean for marketing teams in practice?
For marketing teams, roles shift from production to steering. Four areas of responsibility gain weight, whether your team has five people or fifty.
1. Steering agents
Someone on the team decides which agents handle which tasks, with which data and within which limits. This role looks more like product management than traditional campaign work. Good agent steering starts with precise briefs, not prompts.
2. Owning approvals
Human-in-the-loop is a working principle in which people review and approve at defined points in an automated process. In marketing, that covers brand presentation, legal claims and sensitive audiences. Approval rights belong to named people, not to the tool.
3. Deciding strategy and positioning
Agents can model options, but they don't choose a direction. Deciding what the brand stands for and which motive it addresses remains a human call. That's where ambition lives: the courage not to follow the average.
4. Building consumer intelligence
Anyone steering agents needs reliable knowledge of what drives consumers. That includes behavioral segmentation, which groups people by motives, triggers and behavior, whereas demographic segmentation groups them by age, income or location. Volto AI delivers this knowledge from real consumer voices instead of assumptions.
How do you bring employees along through the AI shift?
Employees embrace the change when they are enabled, experience early wins of their own and know what is happening to their role. Fear grows where leadership stays silent.
The problem in many companies: AI launches as a tool rollout, not as a change in how work gets done. The following sequence starts with people and makes the value measurable:
Be transparent. Say openly which tasks agents will take over and which they won't. Spell out how roles will change. If you can rule out job cuts, say so. If you can't, say that too.
Enable before you roll out. Give your team time and space to experiment with agents before targets are attached. Base training on real tasks from their own work, not on demo cases.
Start small pilots with a clear question. Pick a use case with manageable risk, such as ad variants for one campaign. Decide upfront which decision the pilot should improve.
Bring in consumer data. This is where Volto data comes in. Consumer Resonance is the measure, tracked by Volto AI, of how strongly a message, product or brand triggers approval, rejection and willingness to act among real consumers. Demand Graphs are data models built by Volto AI that map which needs, motives and triggers drive demand in individual consumer segments. With both, you check whether a positioning holds before agents scale it.
Measure and share results. Define metrics before launch, for example time from brief to approval, the share of agent outputs approved without edits, and the Consumer Resonance of messages before and after the pilot. Share the results with the whole team, including the failures.
This sequence turns fear into ownership. Employees who have steered a pilot themselves tend to see AI as a tool for their work rather than a replacement.
What role does Volto AI play in agentic marketing?
Volto provides the consumer intelligence layer for strategy and operationalization, advises on implementation, analyzes where consumer data can be used in marketing, and helps build an agentic architecture for marketing.
Agentic Customer Intelligence is a term coined by Volto AI for customer knowledge drawn from real consumer voices that AI agents can access directly (for example via MCP) and use for marketing decisions. The Model Context Protocol (MCP) is an open standard that lets AI agents access external data sources and tools. Volto makes Consumer Resonance and Demand Graphs available via MCP, so agents work from real consumer data instead of plausibility.
The foundation is qualitative-at-scale research. Qualitative at Scale Research is qualitative market research with thousands of real respondents whose open-ended answers are analyzed with AI in hours instead of weeks. This lets Volto AI combine the depth of qualitative interviews with the speed agentic workflows demand. How brands and agencies align their entire setup around this is covered in Volto's article “Agentic marketing: how brands and agencies are rebuilding their setup for AI agents”.
Frequently asked questions
Will AI replace marketers?
No. AI mainly replaces tasks, not entire marketing jobs. AI agents take over execution such as copy variants, campaign delivery and reporting. People own brand direction, positioning, approvals and the understanding of real consumers. Some routine activities disappear, while new roles such as agent steering and consumer intelligence emerge. Marketers who reinvent their work gain influence rather than lose it.
Why do so many AI projects in marketing fail?
Many AI projects fail for lack of direction, not because of the technology. In June 2025, Gartner predicted that over 40 percent of agentic AI projects will be canceled by the end of 2027, partly due to escalating costs and unclear business value [3]. Typical causes in marketing are missing goals, no predefined metrics and agents that optimize for the average without real consumer knowledge.
Which tasks do AI agents take over in marketing?
In marketing, AI agents mainly take over repeatable execution: creating content variants, delivering to audiences, shifting budgets, setting up and analyzing A/B tests and generating reports. People remain accountable for brand voice, positioning, approvals and the interpretation of consumer data. The line is drawn wherever judgment about the brand, risk and people is required.
What new roles does AI create in marketing teams?
Four areas of responsibility become more important: agent steering (which agent handles which task with which data), approval ownership based on the human-in-the-loop principle, strategy and positioning, and consumer intelligence. The last role ensures that agents work from real knowledge of the audience's motives and triggers rather than from plausible assumptions.
How do you bring employees along during AI adoption?
Employees come along when three things come together: transparency about changing roles, enablement on real tasks before the rollout, and small pilots with a clear question and measurable metrics. Sharing results openly, including failures, builds trust. Fear grows where leadership stays silent or frames AI purely as a cost-cutting program.
What does Volto AI do in agentic marketing?
Volto provides the consumer intelligence layer for strategy and operationalization, advises on implementation, analyzes where consumer data can be used in marketing, and helps build an agentic architecture for marketing. The foundation is Consumer Resonance and Demand Graphs built from real consumer voices, which AI agents can access directly via MCP.
Conclusion
AI won't replace marketers. It replaces execution, and it exposes where direction is missing. Data from McKinsey, Gartner and MIT shows the same pattern: adoption is easy, but impact only comes with clear goals, new roles and real consumer knowledge.
For marketing teams, the scarce resource is vision, ambition and judgment. Volto AI provides the consumer intelligence layer for exactly that: Consumer Resonance and Demand Graphs from real consumer voices, usable for strategy and directly by AI agents via MCP. Volto also advises on implementation and helps build an agentic architecture for marketing.
The technology is available to everyone. The difference is whether you shape the change or wait it out. Which side do you want to be on?
Want to know where consumer data gives your marketing the biggest lever? Talk to Volto AI about your consumer intelligence layer and your first agentic pilot.
Try the MCP integration
About the author
Julia Kirsch, CEO & Co-Founder, Volto AI. Founded Volto AI in Berlin in 2024 and leads strategy and client work in qualitative research, consumer resonance and agentic marketing.
Published: 15 October 2026 · Last updated: 15 October 2026
Sources
McKinsey & Company (2025): The State of AI: Global Survey 2025
Silicon Canals (2025): McKinsey's 2025 global AI survey: 88% of organizations now use AI in at least one function, up from 78% — but most are still stuck in pilot mode, and only a minority can point to any real impact on profit
Gartner (2025): Gartner Predicts Over 40% of Agentic AI Projects Will Be Canceled by End of 2027
Fortune (2025): MIT report: 95% of generative AI pilots at companies are failing
World Economic Forum (2025): Future of Jobs Report 2025: 78 Million New Job Opportunities by 2030 but Urgent Upskilling Needed to Prepare Workforces
HR Dive (2025): Nearly 4 in 10 companies will replace workers with AI by 2026, survey shows
Volto AI (2026): Wechselverhalten 2026
Volto AI (2026): Dynamische Tarife 2026
Bitkom (2026): Erstmals nutzt die Mehrheit der Unternehmen KI
TU Darmstadt (2026): Wer KI versteht, hat mehr Angst vor Jobverlust (KI-Monitor 2026)
ZDFheute (2026): KI-Firma DeepL baut ein Viertel der Jobs ab
Stanford Digital Economy Lab (2026): Canaries in the Coal Mine? Six Facts about the Recent Employment Effects of Artificial Intelligence
Y Combinator (2025): Alexandr Wang: "This is a Once-in-a-Civilization Opportunity"
