2026 AI and Marketing Predictions

Published January 2-12, 2026

2025 – 1

A year ago, I had compiled predictions for 2025. This is a similar exercise: a lookback at the year that was and the year to come. Stepping back from the daily news flow helps give a perspective on how far we have come and the road ahead.

TIME magazine named “The Architects of AI” as its 2025 Person of the Year. “This is the story of how AI changed our world in 2025, in new and exciting and sometimes frightening ways. It is the story of how [Jensen] Huang and other tech titans grabbed the wheel of history, developing technology and making decisions that are reshaping the information landscape, the climate, and our livelihoods. Racing both beside and against each other, they placed multibillion-dollar bets on one of the biggest physical infrastructure projects of all time. They reoriented government policy, altered geopolitical rivalries, and brought robots into homes. AI emerged as arguably the most consequential tool in great-power competition since the advent of nuclear weapons.” More: “This was the year when artificial intelligence’s full potential roared into view, and when it became clear that there will be no turning back or opting out. Whatever the question was, AI was the answer. We saw it accelerate medical research and productivity, and seem to make the impossible possible. It was hard to read or watch anything without being confronted with news about the rapid advancement of a technology and the people driving it. Those stories unleashed a million debates about how disruptive AI would be for our lives. No business leader could talk about the future without invoking the impact of this technological revolution. No parent or teacher could ignore how their teenager or student was using it.”

Andrej Karpathy reviewed the progress in LLMs in 2025. “2025 was an exciting and mildly surprising year of LLMs. LLMs are emerging as a new kind of intelligence, simultaneously a lot smarter than I expected and a lot dumber than I expected. In any case they are extremely useful and I don’t think the industry has realized anywhere near 10% of their potential even at present capability. Meanwhile, there are so many ideas to try and conceptually the field feels wide open.”

Chinese AI in 2025: “A year for the history books for the Chinese AI beat. We began the year astonished by DeepSeek’s frontier model, and are ending in December with Chinese open models like Qwen powering Silicon Valley’s startup gold rush.”

I began by asking the AIs themselves about how they say the big developments of 2025.

Perplexity:

2025 marked a pivotal year for AI, with breakthroughs in reasoning models, multimodal systems, and agentic architectures accelerating real-world deployment across industries. Enterprises scaled production-grade reasoning AI, while hardware and software innovations addressed compute bottlenecks. Marketing applications emerged prominently, leveraging these for hyper-personalized campaigns and autonomous content creation.

Google’s Gemini 3 launch in November set new benchmarks, scoring 91.9% on GPQA Diamond for PhD-level reasoning and leading LMArena at 1501 Elo, enabling sustained complex interactions [ from prior].
Microsoft released proprietary models MAI-Voice-1 for ultra-fast audio generation (one minute in under a second) and MAI-1 Preview as a foundational LLM, marking its independence from OpenAI.
OpenAI’s GPT-4.5 research preview in February introduced its largest model, advancing multimodal reasoning and efficiency.
Tesla’s Optimus robot achieved enhanced dexterity for factory tasks, while Purdue’s RAPTOR AI hit 97.6% accuracy in semiconductor defect detection [ from prior].
Reasoning AI entered enterprise production, with banks like Morgan Stanley deploying fine-tuned LLMs for compliance, risk, and decision support, reducing false positives in financial crime detection.
Multimodal AI systems integrated text, images, video, and audio for holistic understanding, powering healthcare diagnostics and content generation.
Autonomous AI agents made real-time decisions in supply chains, diagnostics, and self-driving vehicles, outperforming humans in optimized scenarios.
Quantum AI progressed with algorithms solving optimization in seconds for drug discovery and logistics, led by IBM and Google.
In B2C martech, generative AI automated ad creatives and dynamic bidding, while predictive analytics enabled hyper-personalization from real-time signals.
Composable martech stacks with AI agents autonomously handled CRM, lead scoring, and localized campaigns, boosting ROI through unified data.

Gemini:

The year has been defined by the collapse of “AI hype” and the rise of “AI utility.” We moved from chatting with bots to managing Agents.

Below is the in-depth retrospective of the top developments of 2025.

2025 will be remembered as the “Year of the Agent.” The industry pivoted from building bigger brains to building better hands—giving AI the ability to execute tasks, not just generate text.

The Rise of “Agentic AI” (From Chat to Action): The most significant shift was the move from “Prompt & Response” to “Goal & Execution.” In 2025, systems like OpenAI’s “Operator” and Anthropic’s “Computer Use” moved from demos to enterprise deployment. These agents didn’t just write email drafts; they opened the inbox, categorized messages, archived spam, and scheduled meetings autonomously.
Small Language Models (SLMs) Overtook Giants: While 2024 was about “Trillion Parameters,” 2025 was the year of efficiency. Enterprises realized they didn’t need a PhD-level model to summarize meetings. Specialized, fine-tuned SLMs (like Microsoft’s Phi-4 and Apple’s on-device models) proved cheaper, faster, and more private than massive cloud models, running locally on laptops and phones.
Physical AI & Robotics Convergence: The “Brain” finally met the “Body.” 2025 saw the first commercial deployment of humanoid robots (like Tesla Optimus and Figure 02) in actual warehouse workflows, capable of understanding semantic commands (“Clean up that spill”) rather than just hard-coded coordinates.
Sovereign AI Infrastructures: Nations stopped relying solely on US tech giants. In 2025, we saw the launch of “Sovereign Clouds” in the EU, India, and the UAE—state-backed AI infrastructure ensuring that sensitive national data (healthcare, finance) never left the country’s legal jurisdiction.
The “Trough of Disillusionment” Correction: As predicted by Gartner, 2025 was a brutal year for “AI Wrapper” startups. Companies stopped paying for generic AI features. The market consolidated around tools that offered hard ROI (revenue or time saved) rather than just “novelty,” leading to a massive purge of SaaS tools.
AI-Fueled Coding (The “80% Draft”): Software development changed fundamentally. By late 2025, AI was generating over 80% of routine boilerplate code. The role of “Junior Developer” effectively vanished, replaced by “AI Architects” who review and stitch together AI-generated modules.
The End of “SEO” (Rise of AEO): Traditional Search Engine Optimization collapsed as organic traffic from Google dropped 40-60% for informational sites. Content creators pivoted to Answer Engine Optimization (AEO)—structuring data to be cited by Gemini/ChatGPT rather than clicked by humans.
Synthetic Data Supremacy: With human data “tapped out,” 2025 proved that Synthetic Data (data generated by high-quality AI to train smaller AI) actually works. This breakthrough allowed models to improve in reasoning and math without needing more internet scrapes.
Real-Time Voice Translation: The “Babel Fish” moment arrived. Services like Live Translate became seamless enough for cross-language business meetings to happen in real-time without human interpreters, breaking down global language barriers in B2B trade.
Marketing: The “Contextual Carry”: In B2C, the “Segment of One” became real. Marketing platforms began using “Contextual Carry”—if a user asked an Instagram bot about hiking, the brand’s website would automatically re-skin itself to show hiking gear the moment the user clicked through, without them logging in.

2025 – 2

Claude

DeepSeek R1: China’s $6M Wake-Up Call to the West

In January 2025, Chinese company DeepSeek released its R1 model under the MIT License, achieving performance comparable to OpenAI’s o1 on benchmarks like AIME and MATH. The company claims that it trained its V3 model for US$6 million—far less than the US$100 million cost for OpenAI’s GPT-4 in 2023—and using approximately one-tenth the computing power consumed by Meta’s comparable model.

The release rattled markets, causing NVIDIA’s stock to dip 17 percent amid fears that demand for its high-performance GPUs might not be as essential as thought. If the United States does not double down on AI infrastructure, incentivize an open-source environment, and overhaul its export control measures, the next Chinese breakthrough may actually become a Sputnik-level event.

Why it matters: DeepSeek proved that frontier AI capabilities don’t require frontier budgets. It validated reinforcement learning-first approaches and demonstrated that U.S. export controls on chips couldn’t kill innovation—only redirect it.

Reasoning Models Came of Age

Reasoning defined the year, as frontier labs combined reinforcement learning, rubric-based rewards, and verifiable reasoning with novel environments to create models that can plan, reflect, self-correct, and work over increasingly long time horizons.

The duration of tasks that large language models can reliably complete with at least a 50 percent success rate has been doubling every seven months since 2019, according to METR. In 2019, leading models could only manage tasks requiring a few seconds of human effort. In 2025, Anthropic’s Claude 3.7 Sonnet boasts a “time horizon” of 59 minutes, and Claude 4.5 has extended this to more than 30 human hours.

Why it matters: The shift from “next token prediction” to genuine step-by-step reasoning transformed what AI could accomplish—from answering questions to completing complex, multi-hour professional tasks.

GPT-5 and the Frontier Model Race Intensified

OpenAI launched GPT-5 on August 7, 2025, combining reasoning and non-reasoning functionality under a common interface with state-of-the-art performance on various benchmarks. GPT-5 achieved 94.6% on AIME 2025 without tools, 74.9% on SWE-bench Verified, and is approximately 45% less likely to hallucinate than GPT-4o.

Google released Gemini 3 in November, calling it “AI for a new era of intelligence” and the best model in the world for multimodal understanding—its most powerful agentic and vibe coding model to date.

OpenAI declared a “code red” to accelerate GPT-5.2’s release by December 9, 2025, countering Google’s Gemini 3, which topped leaderboards and wowed Sam Altman himself.

Why it matters: The major labs are now releasing frontier models every few months, not years. Competition is driving rapid capability gains—but also raising questions about safety and sustainability.

AI Became a Scientific Collaborator

AI is becoming a scientific collaborator, with systems like DeepMind’s Co-Scientist and Stanford’s Virtual Lab autonomously generating, testing, and validating hypotheses. In biology, Profluent’s ProGen3 showed that scaling laws now apply to proteins too.

In 2026, AI won’t just summarize papers, answer questions and write reports—it will actively join the process of discovery in physics, chemistry and biology. “AI will generate hypotheses, use tools and apps that control scientific experiments, and collaborate with both human and AI research colleagues,” says Peter Lee, president of Microsoft Research.

Google marked the 5-year anniversary of AlphaFold cracking the protein folding problem. The profound scientific and societal value of this work was recognized in 2024 with the Nobel Prize in Chemistry.

Why it matters: AI is moving from being a tool scientists use to being a collaborator that proposes hypotheses and designs experiments—potentially accelerating discovery by orders of magnitude.

Agentic AI Moved from Concept to Production

By late 2025, 52% of executives report their organizations have already deployed AI agents, with 88% achieving measurable ROI. Agentic commerce—shopping powered by autonomous AI agents—is transforming from experimental technology into competitive necessity.

Traffic to US retail sites from GenAI browsers and chat services increased 4,700% year-over-year in July 2025, according to Adobe. Customers arriving via AI agents are 10% more engaged than traditional visitors, reaching retailers further down the sales funnel with a stronger intent to purchase.

OpenAI announced an Agentic Commerce Protocol, codeveloped with Stripe, which allows users to complete purchases within ChatGPT without leaving the chat. Shopify is developing an agentic shopping infrastructure that allows agents to tap into its catalog and build carts across merchants. Amazon, Google, PayPal, Mastercard, and others are also developing agentic shopping services.

Why it matters: AI agents aren’t just answering questions—they’re booking travel, comparing prices, negotiating deals, and completing purchases. The intermediary layer between brands and consumers is being rewritten.

Enterprise AI Adoption Crossed the Chasm

Commercial traction accelerated sharply. Forty-four percent of U.S. businesses now pay for AI tools (up from 5% in 2023), average contracts reached $530,000, and AI-first startups grew 1.5× faster than peers, according to Ramp and Standard Metrics.

Our inaugural AI Practitioner Survey, with over 1,200 respondents, shows that 95% of professionals now use AI at work or home, 76% pay for AI tools out of pocket, and most report sustained productivity gains—evidence that real adoption has gone mainstream.

Software development activity on GitHub reached new levels in 2025. Each month, developers merged 43 million pull requests—a 23% increase from the prior year. The annual number of commits jumped 25% year-over-year to 1 billion.

Why it matters: AI moved from pilot projects to production deployments. The question shifted from “should we use AI?” to “how fast can we scale it?”

Small Language Models Proved Bigger Isn’t Always Better

2025 saw a pivot toward efficient small language models (SLMs). Cost pressure, latency requirements, and privacy demands forced enterprises to rethink the assumption that bigger is better. Models with 3B to 15B parameters became workhorses of the year, often costing under $0.0001 per request to run.

Fine-tuned small language models are built for specific purposes and trained on focused data, providing high accuracy for their specialized tasks. They’re breaking the old adage: “Between good, cheap and fast, choose two.” These SLMs can provide all three benefits, often performing comparatively with larger models in accuracy while outperforming on speed and costs.

Domain-specific language models (DSLMs) fill the gap where generic LLMs fall short for specialized tasks. By 2028, Gartner predicts that over half of the GenAI models used by enterprises will be domain-specific.

Why it matters: The “race to the biggest model” gave way to a focus on efficiency, specialization, and practical deployment. Enterprises realized they needed models optimized for their specific use cases, not general-purpose giants.

AI Search Began Eating Traditional Search

ChatGPT Search now handles over a billion searches weekly. It shows links by design and is adding product-style results. The discovery pie is growing—but it’s being sliced differently.

ChatGPT now owns 4.3% of search share—a jaw-dropping rise for a platform that wasn’t even on the radar a few years ago, and it’s racked up over 400 million weekly users.

AI platforms are expected to drive more website visits than traditional search engines in the next three years. Traffic from large language models rose from about 17,000 to 107,000 sessions comparing January-May 2024 with the same period in 2025.

Why it matters: Brand discovery is being rewired. When consumers ask ChatGPT “what’s the best X?” instead of Googling, decades of SEO investment become less relevant. Brands must now optimize for AI answers, not just search rankings.

Generative AI Transformed Marketing (from “Walk” to “Run”)

72% of marketers identified GenAI as the most important consumer trend heading into H2 2025—a 15-point increase from late 2024. GenAI ranked first across every vertical surveyed, from telecom and retail to travel and financial services.

90% of marketers now use AI for text-based tasks, with the most common applications being idea generation (90%), draft creation (89%), and headline writing (86%). ChatGPT dominates at 90% usage, followed by Google Gemini at 51% and Claude at 33%.

GenAI is now being used for copywriting by 34% of marketers, image generation by 25%, and creative versioning by 25%. Website development usage jumped nearly 70% since last year.

Why it matters: Marketing teams moved from experimenting with AI to deploying it across content creation, personalization, and campaign optimization at scale. The productivity multiplier became real.

The Martech Stack Exploded (and Imploded Simultaneously)

The martech landscape grew 9% to 15,384 solutions, with AI-native products continuing to blossom while the previous generation consolidated.

62.1% of respondents now use more tools than two years ago, with generative AI tools used by 68.6% of organizations—making them the 6th most popular martech tool category in just two years.

Popular AI assistants like ChatGPT, Claude, and Gemini are creating software programs behind the scenes to do users’ bidding without them even knowing software was built. The result: not millions but billions of custom software programs proliferating across digital operations.

Why it matters: The “hypertail” of custom AI-powered apps emerged alongside commercial tools. Marketers can now build their own solutions without engineering teams, fundamentally changing the build-vs-buy equation.

2025 – 3

ChatGPT

AI agents moved from demos to early production: Enterprises started deploying “agentic” systems that can take multi-step actions (not just answer prompts), especially in ops, support, and internal productivity workflows.
Multimodal became default, not exotic: Models that natively work across text + images (and increasingly audio/video) became mainstream in products, enabling richer UX and more robust real-world understanding.
Inference economics (cost to run AI) fell sharply: The 2025 Stanford AI Index added deeper tracking of inference cost and hardware trends—highlighting how rapidly “serving” models is getting cheaper, driving broader deployment.
Hardware roadmaps accelerated around inference and “AI factories”: AI infrastructure focus shifted from “can we train bigger?” to “can we run it cheaply at massive scale?”, with new GPU/platform announcements explicitly tuned for inference and huge contexts.
Long-context capabilities pushed into the million-token era (and reality checks followed): Several frontier systems expanded context windows dramatically, enabling full-codebase / long-document workflows, while the industry simultaneously got more honest about what long context does and doesn’t solve. NVIDIA Newsroom+1
Smaller, domain-specific models gained momentum: Companies increasingly mixed frontier models with specialised or smaller models for privacy, latency, and cost control—especially for “repeatable” enterprise tasks.
Governance and regulation stopped being theoretical: The EU AI Act moved through concrete enforcement milestones in 2025 (e.g., prohibited practices + AI literacy requirements in Feb 2025; general-purpose AI obligations from Aug 2025), forcing real compliance programmes.
Security and provenance became first-class concerns: As synthetic content and AI-enabled cyber risk grew, “digital provenance” and AI security platforms rose as priority technology areas rather than niche topics.
A capital and M&A wave built around AI infrastructure and apps: Deal flow and investment increasingly concentrated on “AI-native” products, compute/inference supply chains, and distribution—reflecting a land-grab mindset.
Marketing-adjacent: AI use surged in marketing & sales functions: Survey data shows AI is widely used across business functions, with marketing and sales among the most common areas of adoption—content generation, targeting, support, analytics, and experimentation.

A few slides from Benedict Evans capture the scale of AI growth:

Ahead – 1

Let’s begin with the predictions from the AIs.

Perplexity

AI will pivot to agentic systems and enterprise ROI, with multimodal and reasoning capabilities standardizing across applications. Marketing predictions emphasize autonomous growth engines amid consumer AI adoption.

Agentic AI proliferates, appearing in 40% of enterprise apps for multi-step task execution, up from under 5%.
Multimodal models become ubiquitous, processing all data types for robotics, IoT, and healthcare.
Near-infinite context windows enable coherent long-term agent interactions.
Synthetic data scales training while preserving privacy.
Domain-specific models dominate finance and healthcare.
Early AGI prototypes emerge in research labs.
AI reasoning platforms optimize enterprise performance with custom silicon and efficacy metrics.
Quantum AI hits practical applications in optimization.
In martech, agentic systems orchestrate full B2C campaigns with GEO for AI search dominance.
Hybrid human-AI teams in marketing demand content engineers for privacy-first personalization.

Gemini

As we enter 2026, the focus shifts from “Adoption” to “Autonomy” and “Accountability.”

The “Agent-to-Agent” Economy (A2A): The biggest commerce trend of 2026 won’t be B2B or B2C, but A2A. Your personal AI assistant will negotiate with a travel company’s booking agent API to find the best flight price. Marketing will shift to optimizing for “Machine Customers” rather than human eyeballs.
Self-Verifying Agents (Solving Hallucination): We will see the deployment of “Critic Models”—secondary AI systems whose only job is to fact-check the primary AI before it executes a task. This “internal monologue” will finally make AI reliable enough for banking and healthcare.
The “Chief Integration Officer” (CIO 2.0): The CIO role will evolve into the Chief Integration Officer. Their job will no longer be managing hardware, but orchestrating “swarms” of AI agents and ensuring they play nice with human employees (and don’t crash the network).
“Zero-Click” Commerce: Consumers will stop browsing catalogs. You will tell your phone, “Order a birthday gift for my mom, under $50, she likes gardening,” and the AI will handle the selection, purchase, and delivery tracking in the background.
The “Authenticity Premium” (Marketing): As AI content floods the web, “Human-Made” will become a luxury label. Brands will use cryptographic watermarks to prove their customer support or creative ads are 100% human, charging a premium for “Real Reality.”
Preemptive Cybersecurity: We will move from “Reaction” to “Preemption.” AI security systems will patch vulnerabilities before they are exploited, engaging in a constant, invisible war with AI malware bots 24/7.
The Death of the “App Store”: We will see the decline of standalone apps. Instead of opening “Uber” or “DoorDash,” your OS-level AI will simply interface with their APIs directly. The “Interface” is dead; the “Service” remains.
Workforce: The “Up-Leveling” Crisis: A skills gap will emerge where companies have plenty of “Senior” capability (AI + Humans) but no way to train “Juniors,” because the entry-level tasks used for training are now fully automated. 2026 will see new “Apprenticeship” models to solve this.
Strict Liability Legislation: Governments will pass laws making companies financially liable for their Agents’ actions. If your AI booking agent accidentally buys 100 tickets instead of 1, you (or the software vendor) are on the hook, not the “glitch.”
Invisible AI (Ambient Intelligence): Hardware will disappear. Wearables (glasses, pins) will perform “Ambient Computing,” watching what you do and offering help (e.g., pulling up a LinkedIn profile of the person you just shook hands with) without you asking.

Ahead – 2

Claude

2025 was about capability breakthroughs: reasoning models, efficient training (DeepSeek), frontier competition (GPT-5, Gemini 3, Claude 4.5), and AI agents moving from demos to deployments. 2026 will be about operationalization: proving ROI, scaling agents across enterprise workflows, rewiring discovery (search, commerce, marketing), and the inevitable correction as hype meets reality.

Agentic Commerce Becomes the Dominant Shopping Paradigm

The year 2025 will likely be the last consumers shop as they do now. Agentic AI is reshaping commerce by making shopping faster, smarter, and effortless. By next holiday season, most shopping journeys will begin, evolve, or end with AI agents.

McKinsey estimates that agentic commerce could influence up to $3 to $5 trillion annually in global retail sales by 2030.

By 2028, AI agents will command $15 trillion in B2B purchases, according to Gartner. Autonomous buying systems, machine-to-machine negotiation, and data-verification frameworks are moving into the mainstream.

What to watch: The brands that optimize for agent discoverability—clean data, structured catalogs, seamless APIs—will win. Traditional SEO and brand advertising will matter less when an AI agent is deciding what to recommend.

AI Moves from Proof-of-Concept to Proof-of-Impact

2026 will be the moment to move from proof-of-concept to proof-of-impact, ensuring AI drives measurable outcomes, trust, and collaboration at scale, whilst laying the foundations for larger-scale transformation to follow.

There is—rightfully—little patience for “exploratory” AI investments. Each dollar spent should fuel measurable outcomes that accelerate business value. We expect that to change in 2026. We now know what good agentic AI looks like—it has proof points like benchmarks that track value that matters to the business.

What to watch: CFOs will demand ROI metrics, not innovation theater. Organizations that can demonstrate P&L impact from AI will accelerate; those still “experimenting” will face budget cuts.

40% of Enterprise Apps Will Use Task-Specific AI Agents

Gartner predicts that 40% of enterprise applications will leverage task-specific AI agents by 2026, compared to less than 5% in 2025.

Hyper-automation in 2026 will be driven by advanced AI, process mining and cross-platform orchestration—automating complex, end-to-end workflows with minimal human input.

What to watch: The agent ecosystem will mature rapidly. Expect standardized protocols for agent-to-agent communication, shared memory frameworks, and governance tools for managing fleets of AI agents.

“Search Everywhere Optimization” Replaces Traditional SEO

Search Everywhere Optimization will replace traditional SEO as the dominant visibility strategy. As AI search tools like ChatGPT, Perplexity, and Gemini continue to gain traction—and platforms like TikTok, Amazon, and YouTube evolve into primary search destinations—discovery will no longer revolve around a single search engine.

Discovery has fractured far beyond the ten blue links. Users now bounce between TikTok, Reddit, YouTube, ChatGPT, Gemini, and AI assistants before ever reaching a website. The content performing best in 2026 is the kind AI can’t easily imitate: opinionated commentary, first-hand experience, data-rich insights, and multimedia storytelling.

What to watch: Brands will need to optimize for AI citations, not just rankings. Structured data, entity recognition, and “answerability” become the new currency of visibility.

Domain-Specific Language Models Outpace General-Purpose LLMs in Enterprise

CIOs and CEOs are demanding more business value from AI, but generic large language models often fall short for specialized tasks. Domain-specific language models fill this gap with higher accuracy, lower costs, and better compliance. By 2028, Gartner predicts that over half of the GenAI models used by enterprises will be domain-specific.

These are specialized AI models trained on the vocabulary, rules, and operational context of a particular industry or function. Instead of trying to be universal, they specialize—delivering higher accuracy, reducing ambiguity, and staying compliant with domain standards.

What to watch: Vertical AI companies will emerge in every industry—legal, healthcare, finance, manufacturing—with models that outperform GPT-5 on industry-specific tasks.

AI Becomes a True Scientific Collaborator

In 2026, AI won’t just summarize papers, answer questions and write reports—it will actively join the process of discovery in physics, chemistry and biology. AI will generate hypotheses, use tools and apps that control scientific experiments, and collaborate with both human and AI research colleagues.

Examples include MatterGen and MatterSim, AI foundation models that help create new materials and simulate how they will perform, accelerating materials discovery for innovations like carbon capture and high-performance batteries for clean energy.

What to watch: Expect breakthroughs in drug discovery, materials science, and climate modeling as AI moves from analysis to active experimentation.

The AI Bubble Shows Signs of Correction

After several years of frenetic investment and breathless expectations, signs point to a necessary correction in the AI marketplace. Venture-backed startups that promised transformative capabilities without sustainable business models will feel the pressure first. This “pop” will not signal the end of AI—far from it. Instead, the market will mature as weaker players fold, stronger innovators expand, and investors prioritize practical value over hype.

The race to build massive data centers has accelerated at unprecedented speed. As efficiency improves and organizations shift from training to inference, pockets of overcapacity may emerge.

What to watch: Consolidation in the AI startup space, more rigorous due diligence from investors, and a flight to quality over hype. The survivors will be those with real revenue, not just impressive demos.

Critical Thinking Skills Atrophy Forces “AI-Free” Assessments

Through 2026, atrophy of critical-thinking skills due to GenAI use will push 50% of global organizations to require “AI-free” skills assessments. As automation accelerates, the ability to think independently and creatively will become both increasingly rare—and increasingly valuable.

By 2027, 75% of hiring processes will test AI proficiency. Meanwhile, 50% of global organizations will require “AI-free” assessments by 2026 to ensure candidates can demonstrate independent reasoning and critical-thinking skills without machine assistance.

What to watch: A bifurcation in talent markets: “AI-augmented” roles that maximize human-AI collaboration, and “AI-independent” roles where human judgment is the premium skill.

A Brand’s AI Becomes Its Identity

“By 2026, brands won’t be defined by logos or slogans; they will be defined by their AI. These customizable agents will become the ultimate brand ambassadors: smart, personalized, and continuously evolving with every exchange. The brands that win will be the ones whose AI delivers a consistently exceptional experience.”

“With popular consumer chatbots getting more agentic by the day, users are already growing accustomed to digital interactions informed by history, preferences, and personalized context. In contrast, generic experiences are starting to feel more and more broken.”

What to watch: Brand equity will increasingly be measured by the quality of AI interactions. Customer experience teams will merge with AI development teams.

AI Within Search Becomes 3X Greater Than Standalone AI Tools

In 2026, daily usage of AI within search will be three times greater than any standalone AI tool—reshaping how people can discover information.

Hardware heats up: inference—the running of AI models—will make up two-thirds of AI compute by 2026. Nearly $100 billion will be invested globally in sovereign AI compute in 2026.

What to watch: The battle for AI-integrated search will intensify. Google’s AI Overviews, Bing Copilot, and ChatGPT Search will compete for the default discovery layer. The winner controls the gateway to consumer attention.

Ahead – 3

ChatGPT

Agents become the primary interface for “getting work done”: Expect more users to delegate tasks to agentic systems that plan + execute across tools, rather than interacting via single prompts. Multi-agent systems are explicitly highlighted as a key strategic trend.
“AI-native” platforms replace bolt-on copilots: Software will increasingly be rebuilt around AI-first workflows (orchestration, memory, tool-use, evaluation), not just “add an AI button.”
Domain-specific models proliferate: Many organisations will standardise a portfolio approach: frontier model(s) + domain models + small on-device models for privacy/latency.
On-device and edge AI expands materially: Running capable models locally (PC/phone) grows fast for privacy, cost, offline use, and responsiveness—pushing hybrid architectures (device + cloud).
Compute becomes even more geopolitically constrained: Supply, export controls, tariffs, and national strategies will keep shaping what chips and clusters can be deployed where—and at what cost.
Security posture shifts to “preemptive” and AI-specific defence: More enterprises adopt AI security platforms and provenance tooling to handle model risk, data leakage, deepfakes, and AI-enabled attacks.
Physical AI keeps moving from pilots to operational rollouts: Robotics, drones, and industrial automation get pulled forward by better perception + planning, and by labour/safety economics.
Regulation spreads from “principles” to enforcement: Compliance will move from policy documents to audits, technical controls, vendor clauses, and operational monitoring—especially in the EU as timelines progress.
Marketing: “Agent intermediated buying” changes discovery: Gartner explicitly predicts buying flows increasingly mediated by AI agents, shifting optimisation from human-facing SEO/PPC to making products machine-readable and optimised for agent selection.
Marketing/commerce: agentic commerce reshapes CX and loyalty: Expect more brands to prepare for consumers using agents to research, compare, and transact—forcing changes in merchandising, loyalty value propositions, and customer experience design.

Here are some more:

Saurav Singh: “Agentic AI Becomes the Digital Workforce. For years, we considered AI as a helper — something that drafts emails, generates code snippets, or answers our questions. But 2026 marks the turning point. AI becomes an agent, not…Imagine an AI system that notices a failing pipeline, diagnoses the root cause, applies the fix, validates the output, and updates the incident ticket — all while you sleep. Or an AI agent in finance that reconciles mismatches, audits numbers, and sends updated reports automatically. This shift isn’t theoretical anymore. Companies have already begun replacing manual operational workflows with autonomous agents that execute tasks similar to those of digital employees. The real question for 2026 isn’t “What can we automate?” It’s “What should the AI handle by default?””

Gaurav Nigam: “The Agentic Economy Will Overtake Traditional Software. 2025 was the year of agentic experimentation; 2026 will be the year of agentic execution. We are witnessing the end of “software as a tool” and the beginning of “software as labor.” The data is compelling: Gartner predicts 40% of enterprise applications will feature task-specific AI agents by year-end 2026, up from less than 5% today. PwC’s research shows 79% of companies are already adopting AI agents, with two-thirds reporting measurable productivity gains. McKinsey’s analysis suggests agentic workflows could automate 60–70% of employee time in sectors like banking and insurance. We will see the first major wave of B2B transactions where an AI agent negotiates, purchases, and executes a contract with another AI agent — no human in the loop.”

Greg Isenberg: “SaaS and agents merge completely in 2026. Every SaaS product becomes an agent platform, and every agent platform builds SaaS features. The ones that don’t adapt die or get bought for pennies.” Fru: “SaaS and Agents Fully Merge. The difference between SaaS and agents collapses. Every product becomes an adaptive agent and every agent quietly builds the features you need in real time. UI-heavy workflows fade as conversational execution becomes the default operating system for work.”

Ricardo Gulko: “Hyper-Personalization at Scale – AI Tailors Customer Experiences in Real Time. AI will enable an unprecedented level of hyper-personalization in customer experience by 2026. Instead of one-size-fits-all service, companies will use AI to dynamically customize each interaction – from product recommendations to pricing – for the individual customer. Advances in real-time data analytics and machine learning mean that every click, purchase, and inquiry can feed into algorithms that instantly adjust the experience. Customers will increasingly expect brands to “know them” and anticipate their needs. Imagine a retail website that rearranges itself on the fly for each shopper, or a banking app that proactively offers tailored financial advice based on a client’s unique spending patterns. This kind of AI-driven personalization drives higher engagement and loyalty, as customers feel understood on a one-to-one basis. In fact, industry research indicates that deeply personalized experiences can boost customer satisfaction significantly and drive up conversion rates. Companies that master this will differentiate their CX – turning data into delight at every touchpoint.”

Andy Markus: “Businesses will begin building on-demand apps, supported by AI agents. Most business applications have traditionally required long development cycles, continued investment, and constant maintenance. AI-fueled coding dramatically accelerates software development cycles, making it feasible for a company to build on-demand apps! Autonomous agents can even independently adapt to new requirements, making redevelopment faster than traditional app cycles. Businesses can respond faster to changing needs, experiment with new solutions, and pivot away from legacy apps that require long-term investment. Traditional apps won’t completely disappear…yet. The ability to launch and iterate on-demand functionality, in a fraction of the time, leveraging agentic AI will enable a more agile and cost-effective choice for immediate business challenges than traditional models in many situations.”

Charles Towers-Clark: “Job Displacement Will Become More Prevalent Due To AI Agents. In September 2025, Saleforce’s CEO announced the reduction of 4000 jobs in customer support due to a greater reliance on AI agents. Arguably, LLMs improve quality, but do not directly replace work. However, as AI agents become better at completing tasks autonomously, more companies will implement them to reduce labor costs in 2026.”

Ahead – 4

Pratik Bhadra: “Welcome to “Agentic Marketing.” This marks the next leap, where autonomous AI agents transition from executing simple tasks to managing complex workflows. As I’ve written before, this isn’t just a “co-marketer” that offers suggestions; it’s an agentic system that can independently orchestrate journeys, allocate budgets, analyze data and deploy campaigns based on a human marketer’s strategic goals… The core problem with marketing’s “AI 1.0” phase is that we’ve been bolting new technology onto old-world processes. We use GenAI to write an email, but we still send it using 20-year-old retention marketing channels like email and SMS designed for “batch and blast” segmentation. An “agentic” system upends this model. A human marketing leader doesn’t build a 10-step journey flow; instead, they give the AI agent a goal. One example I have used looks like this: “Reengage all dormant customers who purchased in Q4 last year. Your budget is $10,000, the goal is a 15% reactivation rate, and you are not allowed to discount more than 20%.” This transformation has moved us from traditional CRM-based marketing to the intermediate predictive AI-based marketing world we’ve been in for the past decade, and is now transitioning into goal-based autonomous marketing.”

Alex Wang: “AI Moves From “Feature” to Workflow Layer. The highest-performing teams aren’t shipping AI buttons. They’re embedding AI inside the actual process: support routing, financial operations, sales planning, engineering sprints, customer success playbooks. It’s the same theme the Glean AI Transformation report surfaced: AI dies when it’s bolted on; it works when it’s embedded.”

John Chambers: “In 2025, AI went mainstream. In 2026, every employee, across all industries, will use AI regularly. AI is unlike any other tech transition we’ve ever seen – faster, more disruptive, more transformative. This isn’t incremental change; it’s a revolution reshaping even the most traditional verticals. For leaders, this is the moment to break away and emerge as a leader that is redefining the industry as a whole. The choice is clear: disrupt or be disrupted; adopt AI or get left behind. And it doesn’t stop at simply adopting AI – it’s the responsibility of your company to train and empower every employee, from entry-level employees to the C-suite, to leverage it. Those who enable their workforce will lead and those who don’t will lose.”

Bernard Marr: “AI In The Physical World. This trend covers the increasing influence of AI on the physical systems and mechanisms that constitute the world around us. It includes autonomous vehicles, which will undoubtedly become increasingly prevalent, as well as humanoid robotic workers that will take on physical labor in warehouse, construction and healthcare settings, and the web of interconnected devices that makes up the increasingly sprawling “internet of things”. In 2026, AI isn’t just powering apps on our phones and the software we use on PCs. As regulatory and security guardrails mature, it’s sharing our homes, industries and workplaces, becoming a tangible presence in our world, and redefining our interactions and relationships with all forms of technology.”

David Ulevitch: “Building the AI-native industrial base. America is rebuilding the parts of the economy that create real strength. Energy, manufacturing, logistics, and infrastructure are back in focus, and the most important shift is the rise of an industrial base that is truly AI native and software-first. These companies start with simulation, automated design, and AI-driven operations. They are not modernizing the past. They are building what comes next. This is opening major opportunities in advanced energy systems, robotics heavy manufacturing, next-generation mining, biological and enzymatic processes that produce the precursor chemicals every industry depends on, and much more. AI can design cleaner reactors, optimize extraction, engineer better enzymes, and coordinate fleets of autonomous machines with a level of insight no legacy operator can match. The same shift is reshaping the world outside the factory. Autonomous sensors, drones, and modern AI models can now give continuous visibility into ports, rail, power lines, pipelines, military bases, datacenters, and other critical systems that were once too large to manage comprehensively.”

Sapphire Ventures:

The Path Clears for Two $1T+ AI IPOs
A $50B+ AI Software Acquisition Reshapes the Market
AI’s Soaring Power Demand Collides With Energy Constraints
50 AI-Native Companies Hit $250M ARR as Hypergrowth Accelerates
AI Takes Over Music & Lands a Grammy
Open, Small and World Models Gain Significant Market Share
Robotics Adoption Ramps Slowly as Industrial Use Cases Lead
AI Becomes an Even More Critical Driver of Modern Defense Strategy
Cybersecurity x AI – Securing the New Attack Frontier
The AI Bubble Debate Rages On

David Cahn: “My prediction for 2026 is that it will be a tale of two AIs. On the one hand, it will be a year of delays, first in data center buildouts, many of which will fall behind schedule, and second, in the AGI timeline. At the same time, AI adoption will continue its relentless rise. In 2025, startups coined the idea of a “$0 to $100M” club of rapidly scaling AI companies; in 2026, we’ll begin to talk about the “$0 to $1B” club.”” On the second AI: “The Relentless Drive Toward AI Adoption. The best AI startups are moving with extreme efficiency—many are earning north of $1M in revenue per employee. This implies market pull vs. a push sale. Today’s entrepreneurs are building “self-improving” companies—they are themselves using AI agents for functions like legal, recruiting, and sales—creating an ecosystem flywheel effect. AI app companies are also riding a compute cost curve that should drive incremental margin improvement, especially as new data centers come online between now and 2030. Finally, with enterprises facing adoption fatigue on DIY implementations, startups are gaining even more momentum.”

Battery Ventures: “While 2024 was dominated by capital-intensive model training, the center of gravity is shifting to inference. As agentic applications—which autonomously plan and execute complex workflows—come online, they will consume vastly more compute power at runtime. We believe this development will continue to drive revenue as a myriad of new agentic applications hit the market. We are also seeing the walled gardens of closed models challenged and believe open models (like that of Reflection AI* and DeepSeek) will grow in adoption, fueling a more diverse AI ecosystem.”

Bessemer Ventures: “The browser will become AI’s control center. AI has already made its way into the browser through assistants and early agentic tools. The next step will be far more transformative. We predict the browser will evolve from today’s basic AI integrations to a full execution environment where agents run tasks, maintain context across sessions, and coordinate workflows across the apps we use every day. In short, the browser won’t just display the internet. It will run it for you.”

SaaStr list:

50%+ of B2B Sales Teams Will Shrink in Size
AI Agents Will Handle 40–60% of Customer Interactions
“Vibe Coding” Becomes the Default Way to Build Software
The Traditional SaaS Exit Model Breaks Down
AI Gross Margins Rise to SaaS Levels (65–75%)
Customer Support Becomes a Profit Center
Token-Based and Hybrid Pricing Models Become Standard
2026 Becomes the Biggest IPO Year in Tech History
AI-Native Companies Achieve 3–5× Revenue per Employee
The First $1 Trillion AI Company Emerges

Ashu Garg:

Enterprise AI Finally Moves from Pilots to Production
Decision Traces Become the New Data Moat
AI Security Becomes a Board-Level Imperative
SaaS Incumbents Fight Back
Agents Eat E-Commerce
Gemini Overtakes ChatGPT in Consumer Usage
An AI Lab Goes Public
Cursor-Like Interfaces Become the Default

Key Themes – 1

I asked Claude and ChatGPT to distill the predictions into a set of key themes.

“2026 is the year AI stops being a capability and becomes an actor—reshaping how work is done, how buying decisions are made, how brands are experienced, and how trust is enforced.”

The Agentic Shift: From Software as Tools to Software as Labour

Core idea: 2026 is the year AI stops assisting work and starts doing work.

Software has always been a tool humans use. In 2026, software becomes a workforce humans manage. The question shifts from “What can we automate?” to “What should AI handle by default?”

What’s converging:

Agentic AI in 40%+ of enterprise apps (up from <5% in 2025)
Multi-agent systems that plan, execute, and self-correct across tools
SaaS and agents merging completely—every product becomes an adaptive agent
Autonomous workflows replacing human execution in support, finance, ops
“Digital employees” that notice problems, diagnose causes, apply fixes, and update tickets while you sleep

Why this is foundational: This isn’t a feature upgrade—it’s a change in what software is. The implications ripple through pricing models, org structures, and the very definition of “work.”

Agentic Commerce: Machine Customers Replace Human Browsing

Core idea: Buying decisions increasingly happen between machines, not humans.

The consumer doesn’t browse ten tabs. They say: “Find me the best winter coat under $200, my size, ships in two days.” The AI handles scanning, validating reviews, confirming delivery, factoring loyalty perks—and completing the purchase.

What’s converging:

“Zero-click” commerce where AI handles selection, purchase, and tracking
Agent-to-Agent (A2A) transactions: your personal AI negotiates with merchant AIs
Decline of browsing, apps, and traditional funnels
Products marketed to machine customers, not human eyeballs
McKinsey: $3-5 trillion in annual retail sales influenced by agentic commerce by 2030
Gartner: $15 trillion in B2B purchases commanded by AI agents by 2028

Why this breaks everything: Discovery, branding, pricing, loyalty, and attribution all fracture when humans are no longer the primary decision-makers. Decades of SEO, paid media, and brand-building investment lose their edge when an algorithm decides what to recommend.

Discovery Rewired: From SEO to Search Everywhere to Agent Optimization

Core idea: Visibility shifts from ranking on Google to being the answer AI gives.

Discovery has fractured far beyond ten blue links. Consumers bounce between TikTok, Reddit, YouTube, ChatGPT, Perplexity, and Gemini before ever reaching a brand’s website—if they reach it at all.

What’s converging:

“Search Everywhere Optimization” replaces traditional SEO
Generative Engine Optimization (GEO) for AI search dominance
AI citations become the new currency of visibility
Structured data, entity recognition, and “answerability” determine who gets recommended
Content that performs best is what AI can’t easily imitate: opinionated, experiential, data-rich
The death of the “App Store”—OS-level AI interfaces with service APIs directly

Why this matters for marketers: Clean catalogs, consistent metadata, real-time inventory feeds determine AI visibility. Marketing evolves from persuasion to precision—agents care about facts (price, quality, sustainability, fulfillment reliability), not emotional appeals.

The Interface Disappears: AI Becomes the Workflow Layer

Core idea: Users stop clicking software; AI orchestrates systems invisibly.

The highest-performing teams aren’t shipping AI buttons. They’re embedding AI inside actual processes—support routing, financial operations, sales planning, engineering sprints. AI dies when it’s bolted on; it works when it’s embedded.

What’s converging:

“AI-native” platforms replace bolt-on copilots
Browser evolves into AI’s control center and execution environment
Conversational execution replaces UI-heavy workflows
On-demand apps built by agents in real-time
On-device and edge AI expands for privacy, cost, and responsiveness
Ambient intelligence through wearables offering help without being asked

Why this is structural: This changes how software is designed, sold, and monetized. The “interface” dies; the “service” remains. Your OS-level AI simply interfaces with Uber or DoorDash APIs directly—no app required.

From Proof-of-Concept to Proof-of-Impact: The ROI Reckoning

Core idea: 2026 ends AI experimentation and starts AI accountability.

There is—rightfully—little patience for “exploratory” AI investments. Each dollar spent must fuel measurable outcomes. After three years of pilots, tolerance for innovation theater collapses.

What’s converging:

CFO-led ROI scrutiny; P&L impact required
AI budgets tied to outcomes, not experimentation
Hype correction and market consolidation
Weaker players fold; stronger innovators expand
The “$0 to $1B” club emerges—AI startups scaling with extreme efficiency
Data center overcapacity as efficiency improves and focus shifts from training to inference

Why this is inevitable: The AI bubble doesn’t burst—it matures. Investors prioritize practical value over demos. The survivors are those with real revenue, not impressive prototypes.

Trust, Governance & Liability Become Hard Constraints

Core idea: As AI acts autonomously, trust becomes enforceable, not optional.

Once AI can act, errors become financially and legally real. If your AI booking agent accidentally buys 100 tickets instead of 1, someone’s on the hook.

What’s converging:

Self-verifying / “critic” agents that fact-check before executing
Strict liability legislation for AI actions
AI provenance, watermarking, and authenticity verification
Preemptive cybersecurity—AI systems patch vulnerabilities before exploitation
Regulation moves from principles to enforcement (especially EU)
Compliance shifts to audits, technical controls, vendor clauses, operational monitoring

Why this accelerates in 2026: Autonomous action creates autonomous liability. Governments and enterprises can no longer treat AI errors as “glitches”—they’re business decisions with consequences.

Key Themes – 2

Brand Identity Becomes AI Identity

Core idea: Brands are increasingly experienced as AI, not messaging.

By 2026, brands won’t be defined by logos or slogans—they’ll be defined by their AI. These customizable agents become the ultimate brand ambassadors: smart, personalized, and continuously evolving with every exchange.

What’s converging:

Hyper-personalization at scale in real time
Brand AI as the primary customer interface
Loyalty evaluated by machines, not humans
AI-mediated CX becomes brand equity
Customer experience teams merge with AI development teams
Generic experiences feel increasingly “broken” compared to personalized AI interactions

Why this matters: Marketing shifts from campaigns to continuous intelligence. The brands that win are those whose AI delivers a consistently exceptional experience—not those with the best tagline.

The Human Premium: Authenticity, Judgment, and Critical Thinking

Core idea: As AI commoditizes execution, uniquely human capabilities become scarce and valuable.

When AI can write, design, code, and analyze, what’s left for humans? Judgment, taste, experience, and authenticity become the premium skills.

What’s converging:

Critical thinking atrophy forces “AI-free” assessments in 50% of organizations
“Human-Made” becomes a luxury label with cryptographic proof
Content that performs best is opinionated, experiential—what AI can’t imitate
The “Up-Leveling Crisis”: no way to train juniors when entry-level tasks are automated
New “Apprenticeship” models to rebuild the skills pipeline
Bifurcation: “AI-augmented” roles vs. “AI-independent” roles where human judgment is the product

Why this matters: The talent market splits. Some roles maximize human-AI collaboration; others prize human judgment as irreplaceable. Authenticity becomes a brand differentiator and a career strategy.

Domain-Specific AI Wins Over General-Purpose Models

Core idea: The “one model to rule them all” era gives way to specialized, fit-for-purpose AI.

Generic LLMs fall short for specialized tasks. CIOs and CEOs demand business value, and that requires models trained on industry-specific vocabulary, rules, and context.

What’s converging:

Domain-specific language models (DSLMs) dominate finance, healthcare, legal, manufacturing
Small language models (SLMs) deliver “good, cheap, and fast” simultaneously
Portfolio approach: frontier model + domain models + on-device models
Over half of enterprise GenAI models will be domain-specific by 2028
Vertical AI companies emerge in every industry, outperforming GPT-5 on specialized tasks
Open-weight models and composable AI architectures reduce vendor lock-in

Why this is the enterprise reality: Precision beats generality. The winners aren’t the biggest models—they’re the most accurate for the task at hand.

Physical AI, Energy & Infrastructure Become Strategic Bottlenecks

Core idea: AI spills out of software and collides with physical constraints.

AI progress increasingly depends on atoms, not just bits. Energy, chips, data centers, and robotics become the rate-limiting factors—and the geopolitical battlegrounds.

What’s converging:

Robotics and physical AI move from pilots to production (logistics, manufacturing, healthcare)
Energy and compute constraints intensify; AI’s power demand collides with grid capacity
Sovereign AI infrastructure: $100B+ invested globally in 2026
AI-native industrial base emerges (simulation, automated design, AI-driven operations)
Compute becomes geopolitically constrained (export controls, tariffs, national strategies)
Data center buildouts fall behind schedule; pockets of overcapacity emerge

Why this is the long-term axis: The invisible infrastructure of AI—chips, power, cooling, physical automation—determines who can build and deploy at scale. The race for AI supremacy is increasingly a race for energy and manufacturing capacity.

Summary: The 10 Themes at a Glance

#	Theme	One-Line Summary
1	Agentic Shift	Software stops being a tool and becomes a workforce
2	Agentic Commerce	AI agents make buying decisions; humans step back
3	Discovery Rewired	Optimize for machine customers, not just human search
4	Interface Disappears	AI becomes the workflow layer, not a feature
5	ROI Reckoning	Proof-of-impact replaces proof-of-concept
6	Governance Hardens	Trust and liability become legally enforceable
7	Brand = AI	Customer experience is AI experience
8	Human Premium	Authenticity and judgment become scarce assets
9	Domain-Specific Wins	Specialized models beat generic giants
10	Physical Bottlenecks	Energy, chips, and infrastructure constrain AI growth

The Throughline for NeoMarketing

For NeoMarketing, themes 1, 2, 3, and 7 form the core narrative:

Agentic commerce means brands aren’t marketing to humans anymore—they’re marketing to algorithms representing humans
Discovery rewired means the “physics engine for customer behaviour” isn’t a nice-to-have—it’s the only way to anticipate what AI agents will recommend
Brand = AI means hyper-personalization shifts from 1:1 human messaging to 1:1 data optimization
The brand with the cleanest, most structured, most semantically rich data wins when ChatGPT decides what to recommend

This is exactly the NeoMarketing thesis: the collapse of the traditional funnel, the rise of AI-mediated customer journeys, and the imperative to build a “World Model” that predicts customer behaviour before the agent even asks.

My Take (focused on Marketing) – 1

Here is what I had written last year (Part 10):

2025 will mark the convergence of three transformative AI technologies that fundamentally reshape how brands connect with customers. The combination of Agentic AI (through AI Co-Marketers), AI Twins, and true N=1 personalisation will finally solve marketing’s persistent “Not for Me” problem – where generic messaging fails to resonate with individual customers.

AI Twins will evolve beyond basic segmentation to create personal digital companions for each customer. Starting with Adtech Twins built from public data, progressing to Madtech Twins incorporating marketing insights, and culminating in individual Singular Twins (MyTwins), these AI replicas will enable unprecedented understanding of customer needs and preferences.

The AI Co-Marketer will serve as the orchestration layer, using Agentic AI to coordinate across these Twins and create what I call “Generative Journeys” – dynamic customer paths that adapt in real-time like Google Maps recalculating routes. This combination will enable true N=1 personalisation at scale, where every interaction feels personally crafted for each customer.

Most importantly, this trinity of AI innovations will transform marketing from mass communication to individual conversation. Rather than bombarding customers with generic messages, brands will engage in meaningful dialogue through AI-powered personal companions. The result? Higher engagement, better retention, and dramatically lower customer acquisition costs as brands shift from endless acquisition to building lasting, profitable relationships.

This AI-powered transformation of customer engagement lies at the heart of what I call “NeoMarketing” – a revolutionary paradigm that moves beyond mass messaging and repeated acquisition to create genuine one-to-one relationships for lifelong retention at scale.

Last year, I wrote about the convergence of Agentic AI, AI Twins, and N=1 personalisation solving marketing’s persistent “Not for Me” problem. The direction was right; the timing, as always, was optimistic.

Marketing is indeed becoming agentic—but the transformation is unfolding across two parallel tracks: operational automation (Marketing Agents with a Co-Marketer running campaigns) and customer-facing intelligence (AI understanding each individual at unprecedented depth). The broad direction remains: “Department of One for Segment of One.”

This year, I want to go deeper on three forward-looking shifts I’m actively working on—each of which I believe will reshape marketing economics, customer understanding, and channel strategy over the 2026–2028 arc.

Alpha: Value-Based Pricing Becomes the Default Expectation

In 2026, the martech procurement conversation starts to change. For the first time, a meaningful subset of CMOs and CFOs will ask a new question:

“Why are we paying for messages and modules when what we want is outcomes?”

The driver is simple: AI collapses the cost of execution. Copy, creatives, segmentation, variants, basic automation—all approach zero marginal cost. When everybody can generate 50 campaign variants in seconds, the differentiator is no longer production volume. The value migrates from “doing” to “delivering.”

That’s the opening for Alpha pricing—borrowed from finance, where Alpha means returns above the benchmark. Instead of paying for the capacity to send emails, brands pay for the incremental revenue those emails generate. Instead of licensing a CDP based on records stored, pricing reflects the lift in customer lifetime value.

The transition unfolds in three phases:

Phase 1 (2026): Humans-as-a-Service. Growth engineers sit between the brand and the platform, committing to measurable lifts—reactivation rate, repeat purchase, contribution margin, retention. They use AI tools but take accountability for results.
Phase 2 (2027): Agent-Assisted Delivery. AI agents take over operational execution—testing, orchestration, anomaly detection, next-best-actions—reducing cost-to-serve and raising confidence in performance commitments.
Phase 3 (2028): Agent-to-Agent Performance Contracts. The vendor’s agent commits to outcomes, reports transparently, and continuously rebalances tactics to keep performance on track. Fully autonomous, fully accountable.

The prediction: The pricing unit shifts from messages sent and features licensed to retention lift and profit per customer. Vendors who cannot align economics will increasingly be treated as interchangeable utilities.

What to watch in 2026: More contracts with variable components, guarantees, “shared upside” structures, and CFO involvement in martech renewal decisions.

Why this matters for the Great Rebalancing: Alpha pricing aligns incentives between brands and vendors for the first time. When martech is priced on outcomes, the ROI of reducing churn becomes as visible as the ROI of acquiring new customers. The rebalancing from acquisition to retention begins when CFOs can see Alpha in their P&L.

Artificial People: Consumer World Models Make Hyper-Personalisation Real

AI has made “personalisation” cheap. It has not yet made it true.

Most personalisation today is still an elaborate form of rules and propensity scores—constrained by sparse signals, weak understanding, and the fundamental limitation of treating customers as rows in a database rather than as living, deciding, evolving humans.

The next leap comes when brands stop modelling “customers” and start modelling customer behaviour. That’s where Artificial People—what I call consumer world models—becomes the foundational primitive.

A World Model is a physics engine for human behaviour. Just as a self-driving car’s world model simulates road conditions, obstacles, and other drivers, a consumer world model simulates:

Decision dynamics: How this person weighs price vs. convenience vs. brand affinity
Temporal patterns: When they’re receptive vs. resistant to outreach
Context sensitivity: How behaviour shifts across channels, devices, moods, life stages
Influence networks: Who and what shapes their preferences

This enables the industry to move from:

Data → Segments → Campaigns

to:

Models → Predictions → Conversations

Artificial People creates conversable, evolving archetypes that map to individual customers and update continuously. Three things become possible that weren’t before:

Stable meaning from messy signals. Archetypes give coherence to sparse and noisy data—you don’t need perfect information to understand someone.
Real-time intent and timing. The model predicts not just what a customer might want, but when they’re most receptive and why.
A single “customer brain” across channels. Email, onsite, app, WhatsApp, support, and commerce all draw from one evolving representation—no more channel silos with conflicting views of the same person.

The prediction: The martech stack shifts from “event-stream + rules” to “world-model + agents.” CDPs don’t disappear, but they become plumbing. The customer model layer becomes the strategic asset—and the new moat.

What to watch in 2026: Vendors talking less about segmentation and more about “customer model layers,” memory, embeddings, and continuously updating representations.

The collision with agentic commerce: Here’s where it gets interesting. In a world where AI agents shop on behalf of consumers, brands aren’t just marketing to humans—they’re marketing to the models that represent humans. The consumer’s AI agent has its own understanding of their preferences. The brand’s Artificial Person has its own simulation. Marketing becomes a negotiation between world models. The brands that win will be those whose models most accurately simulate customer behaviour—and can “speak” to the customer’s own AI agent in terms it understands.

NeoMails: The Inbox Becomes an Attention and Monetisation Surface

Email has been treated as a cost centre—a channel brands pay to access. Open rates reflect fatigue, not engagement. The inbox is polluted by volume, and every marketer’s response to declining performance is to send more.

In 2026, two forces collide to change this:

The Attention Crisis: As AI agents and AI-generated content explode, human attention becomes scarcer and more expensive. The moments when a human actually reads something become precious.
The Owned Media Realisation: Brands will try harder to reclaim engagement from rented platforms (Meta, Google) by creating repeatable rituals on owned channels. Email is the last direct line to consumers who have explicitly opted in.

NeoMails reimagines email through a simple inversion: instead of brands paying to send emails that customers ignore, brands create emails so valuable that advertisers pay to be included—and customers want to open.

The model:

ZeroCPM Marketing Emails: The brand’s email to its customer costs nothing to send because it’s subsidised by relevant, non-competitive advertising embedded within.
Attention as Currency: The customer’s demonstrated attention—opens, reads, clicks, time spent—becomes a monetisable asset. Brands share that value with customers (through rewards, exclusive content, better offers) or reinvest it in better experiences.
The Inbox as Daily Utility: The strongest email programs will look less like campaigns and more like daily utilities—content, games, recommendations, micro-experiences, interactive moments. They become habit-forming, not interruptive.

The prediction: The winning owned-channel programs will look less like promotional calendars and more like media products with recurring formats and rituals. Email gets reborn as a product, not a channel. And new monetisation structures—not increasing ESP bills—will fund attention creation.

What to watch in 2026: More interactive email, more embedded actions, more inbox-native experiences, more experimentation with ad-supported owned media.

Why this works in an agentic world: As AI agents filter and triage communications on behalf of consumers, only high-value, high-relevance messages will get through. NeoMails are designed to pass that filter—they deliver genuine value, not noise. The brands that master this will be the ones whose emails AI agents allow into the consumer’s attention stream.

My Take (focused on Marketing) – 2

Five More Predictions for Marketing in 2026

Beyond Alpha, Artificial People, and NeoMails, here are five observable predictions that tie together the NeoMarketing thesis:

Marketing Ops Becomes Goal-Based, Not Journey-Based

Marketers will increasingly specify constraints + goals (“reactivate dormant Q4 customers, $10K budget, 15% target, max 20% discount”), and agents will build and continuously adjust the plan. “Journeys” become an audit trail, not a design artifact. The marketer’s job shifts from building flows to setting objectives and guardrails.

The Primary KPI War Shifts from CTR/ROAS to Retention Economics

Expect a visible shift to metrics like repeat rate, churn prevention, margin contribution, and engagement persistence. AI makes top-of-funnel output cheap, but profitable retention remains hard. The CMOs who win budget battles will be those who can speak the language of lifetime value, not campaign performance.

Personalisation Splits into Two Tiers: “Cheap Content” vs. “Earned Relevance”

Most brands will flood customers with AI-generated variants—more emails, more ads, more noise. The winners will be those who combine modelling + timing + restraint to feel human and respectful. Relevance becomes a premium. The “Not for Me” problem doesn’t get solved by more content; it gets solved by better understanding.

“Answerability” Becomes a Core Marketing Function

As discovery fragments across ChatGPT, Perplexity, TikTok, and AI assistants, brands will optimise for being cited and selected by AI interfaces. This means structured data, strong first-party content, clear entity truth, and machine-readable product information. A new discipline emerges—call it GEO (Generative Engine Optimisation) or Agent Optimisation—that sits alongside (and eventually supersedes) traditional SEO.

The Customer Model Layer Becomes the New Moat

Brands will compete not on how many tools they have, but on the quality of their customer representation and learning loop. The strategic asset isn’t the CDP, the ESP, or the analytics platform—it’s the world model that understands each customer and improves with every interaction. This is what Artificial People enables, and it’s what separates brands that truly know their customers from those that merely have their data.

The Great Rebalancing

Together, these shifts—Alpha pricing, consumer world models, inbox-as-attention—mark the beginning of a fundamental rebalancing: from acquisition-led growth to retention-led profitability.

For two decades, digital marketing has been an acquisition arms race. Brands pour money into Google and Meta to capture new customers, then struggle to keep them. CAC rises relentlessly; LTV stagnates. The math only works with venture subsidies or market dominance.

Agentic commerce breaks this model. When AI agents mediate discovery, the lowest-friction, highest-trust option wins—not the one with the biggest ad budget. Retention becomes the moat. Owned channels become profit centres. The brands that keep customers (and their agents) coming back don’t need to acquire as aggressively.

And central to it all is the insight that will define the next era: In an age where AI makes production limitless, the constraint is no longer content or computation. It is human Attention.

The marketers who win will not be those who send more. They will be those who earn more—more attention, more intention, more repeat behaviour, more trust. That’s the heart of NeoMarketing. And 2026 is the year it stops being a vision and starts being a necessity.

Summary: The NeoMarketing Predictions

#	Prediction	Core Shift
1	Alpha Pricing	From messages licensed to outcomes delivered
2	Artificial People	From segments to world models
3	NeoMails	From cost centre to attention surface
4	Goal-Based Ops	From journey design to objective setting
5	Retention KPIs	From ROAS to lifetime value
6	Earned Relevance	From content volume to timing + restraint
7	Answerability	From SEO to agent optimisation
8	Customer Model Moat	From tool stack to understanding layer

Share this:

Like this: