AI Has Left the Chat: Forrester's Three-Layer Framework for 2026 Strategic Planning

See also (wiki): physical-ai-capability-levels · assistive-to-agentic-shift · agentic-ai-governance

Executive Summary

• Forrester’s flagship 2026 emerging-technology framework (Brian Hopkins, April 9, 2026) organizes ten technologies into three layers — Interact, Build, Fuel — and three investment horizons (short-term: 2 years; medium-term: 2–5 years; long-term: 5+ years). The framing headline: “AI has left the chat — literally.”
• Short-term bets (2 years): Agentic Commerce, AI Security and Trust. Medium-term bets (2–5 years): Agentic Software Development, Humanoid Robots (Physical AI). Long-term bets (5+ years): Quantum Computing.
• The most operationally relevant data point: warehouses, factories, and hospitals report 20–50% efficiency improvements from physical AI deployments, with vendor examples including Google Gemini Robotics (RT-2) and Meta V-JEPA.
• Hopkins introduces “Layer Zero Experiences” — an AI-driven intelligence layer that floats above apps and websites, interprets intent, and orchestrates services. OpenAI’s Apps SDK and Google’s agent-to-UI protocol are the first shipping examples. This is the category that quietly disintermediates brand-owned digital surfaces.
• Quantum note for CIO/CISO audiences: Forrester flags Q-day — quantum computers capable of breaking current encryption — as potentially arriving by 2030. Post-quantum cryptography planning belongs on the 2026–2027 CIO roadmap, not the 2029 one.

Why the Three-Layer Framework Matters for Mid-Market CIOs

The Forrester framework is useful because most mid-market 2026 AI planning is still single-dimensional — a list of point tools without a sequencing logic. Hopkins separates ten technologies into three layers that describe where in the stack the technology applies, then cross-cuts each with a time-to-value horizon.

Interact layer — technologies that shape how people (and AI intermediaries acting on their behalf) engage with products and services. This is the layer where brand disintermediation risk is highest. The four Interact technologies: Layer Zero Experiences, Physical AI and Robotics, Autonomous Transportation, and Agentic Commerce.

Build layer — technologies that change how software and AI systems are engineered, orchestrated, and secured. The three Build technologies: Agentic Software Development, Multi-Agent Systems, and AI Security and Trust.

Fuel layer — the infrastructure and foundational capabilities that power everything above. The three Fuel technologies: Frontier Models, AI Supercomputing, and Quantum Computing.

The cross-cut with investment horizons matters because it tells a board what deserves capital now vs. what warrants a tracking position. Short-term bets are where competitors will capture value first. Long-term bets are where the downside of inaction is strategic positioning, not current P&L.

The Four Technologies Hopkins Elaborates

Hopkins spotlights four technologies as “telling the story of 2026.”

1. Layer Zero Experiences (Interact). Hopkins describes this as “an AI-driven intelligence layer that floats above today’s rigid apps and websites.” Instead of customers navigating app-by-app, an AI intermediary interprets intent and composes services across systems. OpenAI’s Apps SDK and Google’s agent-to-UI protocol are the first shipping examples. For a mid-market CMO or e-commerce leader, this is the same discovery-funnel disintermediation risk IBM IBV + Adobe flagged in their April 2026 customer-intent research — where the AI chooses which brand to surface, brand preference gives way to AI-routable product specifications.

2. Physical AI and Robotics (Interact). Hopkins anchors this with a hard data point: warehouses, factories, and hospitals report 20–50% efficiency improvements from physical AI. Vendors named: Google Gemini Robotics (RT-2) — a vision-language-action model that learns tasks from demonstration — and Meta’s V-JEPA world-model architecture. Horizon: 2–4 years to broad enterprise value. This triangulates with BCG’s April 14, 2026 physical-AI framework (five-level capability maturity matrix, <1M to >6M humanoid-unit forecasts by 2030), giving physical-operations CIOs a second institutional voice to calibrate investment pacing.

3. Multi-Agent Systems (Build). Networks of specialized agents that plan, delegate, and execute across complex workflows. Hopkins names customer support, incident triage, and software delivery as the near-term use cases. Horizon: 2–4 years for broad enterprise value; early adoption today. This is the architectural next step beyond single-agent copilots — and the capability most directly validated by Anthropic’s April 2026 Agentic Coding Trends Report, which lists “multi-agent coordination” as one of four CIO 2026 priorities and cites Fountain’s hierarchical multi-agent orchestration delivering 50% faster candidate screening.

4. Quantum Computing (Fuel). Hopkins flags Q-day — quantum computers capable of breaking current public-key encryption — as potentially arriving by 2030. Applications beyond the defensive include portfolio optimization, molecular simulation, and supply chain routing. For mid-market CIOs and CISOs, the actionable 2026 move is not quantum hardware procurement; it is post-quantum cryptography readiness — inventorying cryptographic dependencies, identifying systems using algorithms NIST has flagged for migration, and sequencing vendor conversations about migration paths.

AI Security and Trust as a Short-Term Bet

AI Security and Trust is categorized as a short-term (2-year) bet — not medium-term. Hopkins describes it as “integrated platforms covering prompts, apps, and models,” addressing prompt injection, data leakage, and model theft. The short-term categorization is significant: Forrester is signaling that the market for consolidated AI security platforms (as opposed to point-solution DLP, prompt firewall, and model-theft tools) is consolidating now, not later.

This is consistent with the Pass 453 Forrester CISO AI-driven-future research (DeMartine, April 2026) and the Pass 459 MIT CISR Minimum Viable Governance research — all three institutional voices landing in Q2 2026 pointing CISOs at the same action: consolidate AI-specific security tooling into a single platform decision this year, don’t wait for the integrated category to mature.

What the Framework Reveals About Corporate 2026 Planning Gaps

Three gaps become visible when an executive team maps its current 2026 AI roadmap against Hopkins’s framework.

The Interact-layer gap. Most mid-market companies have 2026 plans for AI inside their digital surfaces — chatbots, recommendation engines, personalization. Few have plans for how their brand will be discovered, compared, and transacted through AI intermediaries that live outside their surfaces. Layer Zero Experiences and Agentic Commerce are both short-to-medium-horizon threats to customer acquisition economics that most 2026 marketing plans do not address.

The Build-layer gap. Companies are adopting coding copilots (assistive AI). Few are planning for multi-agent software delivery — where agents handle ticket triage, first-pass code review, dependency updates, and test generation, coordinated by a human architect rather than directed step-by-step. Forrester’s 2–4-year horizon means the competitive gap opens in 2027–2028 and the companies that retooled engineering operating models in 2026 will be harder to catch.

The Fuel-layer gap. Quantum is the easy one to defer. The harder one is Frontier Models and AI Supercomputing — specifically, the capex trajectory Deloitte’s March 2026 AI Infrastructure Survey (n=515) documents (AI-infrastructure budgets projected to more than triple through 2028; 61% of >$500M-revenue respondents expect 10B+ tokens/month). Mid-market buyers do not build supercomputers, but they negotiate vendor contracts whose pricing trajectories are driven by the capex structure at the scale above them.

Key Data Points

Source credibility: MEDIUM-HIGH. Brian Hopkins’s blog (April 9, 2026) and Forrester press release (April 15, 2026) are analyst-framework publications, not primary-survey research. The 20–50% physical-AI efficiency figure is cited without a sample size or underlying survey source; it is directionally consistent with published physical-AI case studies but cannot be treated as a measured benchmark. All layer assignments, horizon categorizations, and vendor-example pairings are from Forrester’s public-facing publications. The full underlying report “The Top 10 Emerging Technologies In 2026” is paywalled behind a Forrester Decisions subscription; findings above are triangulated from the public blog + press release surface area. Apply Forrester analyst caveat: Hopkins’s Emerging Tech Portfolio practice has commercial interest in Forrester Wave evaluations covering the vendors named; the ten-technology framework is prescriptive analyst judgment, not empirical RCT data.

How This Connects to Existing Evidence

BCG Physical AI (April 14, 2026) anchors the same physical-AI trend with a five-level capability framework and humanoid-robot forecast ranges (<1M to >6M units by 2030). Forrester’s 20–50% efficiency figure is the near-term benefit side of the same story BCG tells in terms of capability maturity.

Anthropic 2026 Agentic Coding Trends Report (April 2026) names multi-agent coordination as one of four CIO priorities and provides enterprise case benchmarks (Fountain 50% faster screening via hierarchical multi-agent orchestration; Rakuten 7-hour autonomous run on 12.5M-line vLLM codebase). Hopkins’s 2–4-year horizon for multi-agent systems aligns with Anthropic’s framing of 2026 as the beginning of the multi-agent scaling phase.

IBM IBV + Adobe customer-intent research (April 2026, n=1,000) and Forrester agentic-payments B2C commerce (Varon, April 9, 2026) together anchor the Agentic Commerce / Layer Zero Experiences short-term bet. IBM names the brand-funnel disintermediation risk; Varon documents what has actually shipped (ChatGPT 900M WAU, Alipay AI Pay 120M weekly transactions, Walmart Instant Checkout 3x lower conversion vs. direct site).

MIT CISR Minimum Viable Governance (van der Meulen et al., March 19, 2026) and Forrester CISO AI-Driven Future (DeMartine, April 2026) triangulate the AI Security and Trust short-term bet. Three institutional voices in Q2 2026 pointing at the same action: consolidate AI security tooling this year.

Deloitte AI Infrastructure Survey (March 30, 2026, n=515 director+ at $500M+ revenue firms) provides the capex-trajectory evidence behind the Fuel layer — AI-infrastructure budgets projected to more than triple through 2028, 61% expecting 10B+ tokens/month.

What This Means for Your Organization

A mid-market CIO using this framework for 2026–2027 strategic planning should run three tests.

1. Map your current roadmap against the three layers. If >70% of planned AI spend lives in the Build layer (copilots, code assistance, internal productivity), the portfolio is under-invested in the Interact layer — where customer acquisition and brand economics are shifting. The gap will not be visible in Q2 2026 but will be expensive to close in Q2 2027 after competitors move first.

2. Assign each technology a defensible horizon position. For each of the ten technologies, the board-level question is not “should we invest?” but “what is our position, and what is the cost of being wrong?” Short-term bets (Agentic Commerce, AI Security and Trust) have active-investment positions. Medium-term bets (Agentic Software Development, Humanoid Robots) have pilot-and-measure positions. Long-term bets (Quantum) have tracking positions with a named owner. The tracking position still requires an owner; otherwise it quietly becomes “no position.”

3. Sequence post-quantum cryptography work now, not in 2029. Q-day by 2030 means the cryptographic inventory, migration pathway planning, and vendor conversations belong in the 2026–2027 CIO workplan. Any long-lived data encrypted today with current algorithms is potentially readable in 2030 if captured and stored. The harvest-now-decrypt-later threat is already operational; the cost of delay is not zero.

The value of Hopkins’s framework is not that it tells you which technology to buy. It tells you which conversations to sequence with your board this year versus next year. If this framework raised questions about how your current 2026 plan maps to the three layers — and whether the investment horizons line up with where your competitive pressure will actually come from — I’d welcome the conversation — brandon@brandonsneider.com.

Sources

1. Forrester, “Forrester’s Top 10 Emerging Technologies For 2026: Beyond Chat,” Brian Hopkins, VP — Emerging Tech Portfolio, April 9, 2026. https://www.forrester.com/blogs/forresters-top-10-emerging-technologies-for-2026-beyond-chat/

2. Forrester Press Release, “Forrester’s Top 10 Emerging Technologies For 2026,” April 15, 2026, Cambridge, Mass. Quote: Sharyn Leaver, Chief Research Officer. https://www.forrester.com/press-newsroom/forresters-top-10-emerging-technologies-for-2026/

3. Forrester, “The Top 10 Emerging Technologies In 2026,” underlying research report (client-gated behind Forrester Decisions subscription).

4. BCG, “How Physical AI Is Reshaping Robotics Today — and What Comes Next,” April 14, 2026. Five-level physical-AI capability framework; humanoid-unit forecasts.

5. Anthropic, “2026 Agentic Coding Trends Report,” April 2026. 60% / 0-20% collaboration paradox; enterprise case benchmarks (TELUS, Zapier, Rakuten, Fountain).

6. IBM IBV + Adobe, “Win the Moment by Mastering Customer Intent,” April 15, 2026 (n=1,000 senior tech and marketing executives). Customer-intent orchestration and brand-funnel disintermediation.

7. Forrester, “Agentic Payments In B2C Commerce: Where We Are Now,” Lily Varon, April 9, 2026. ChatGPT 900M WAU, Walmart Instant Checkout 3x lower conversion.

8. MIT CISR, “Minimum Viable Governance for Generative AI,” van der Meulen, Jewer, Levallet, March 19, 2026. Governance framework for AI deployment.

9. Forrester, “CISOs Have Plenty Of Work To Do In An AI-Driven Future,” Amy DeMartine, April 9, 2026. CISO role redesign for embedded AI.

10. Deloitte Insights, “Enterprise AI Infrastructure Survey: A 2028 Outlook,” March 30, 2026 (n=515 US director-level+ at $500M+ revenue organizations). AI-infrastructure capex trajectory.

11. For context on cross-reference to vendor case studies: METR RCT (experienced developers 19% slower, n=16, July 2025); CMU study (40.7% code complexity increase); Atlan 200-deployment analysis (median +159.8% ROI requires workflow redesign first); Faros data (98% more PRs, same delivery throughput — bottleneck moved from coding to review).

Brandon Sneider | brandon@brandonsneider.com April 2026