What is the current thing?
AI is the current platform wave. Agents are the near-term product shape. Robotics is the next credible frontier. NFTs are the warning label: a real primitive can still carry a speculative consumer story that fizzles.
Wardley mapping separates visibility from maturity. A thing can be very visible and still immature, like agents. It can be mature but less talked about, like cloud. It can be visible and dead as a broad thesis, like retail NFTs.
AI as a general-purpose platform
The broad wave is not chatbots. It is foundation models, cheaper inference, AI infrastructure, software copilots, workflow agents and model-native products.
Robotics / physical AI
Robotics has the right dependency stack: better perception, simulation, cheap inference, more robot data and a real labor/reshoring pull. It is hard because atoms are slower than tokens.
Quantum and synthetic biology
Both can matter enormously, but the Wardley position is still earlier: domain-specific, expensive, talent-constrained and not yet broadly productized.
NFTs and metaverse
The lesson is not that all hype is fake. The lesson is that a meme needs a utility migration path. NFTs kept provenance mechanics; the speculative image market did not.
| Era | Wave | Original claim | Status | What happened |
|---|---|---|---|---|
| 1990s | Internet | Packet networks turn publishing, search and commerce into software. | Scaled | Became the substrate for every later platform wave. |
| 2007-2015 | Mobile | A sensor-rich computer moves into every pocket. | Scaled | Evolved into the customer interface for cloud, social and AI. |
| 2010s | Cloud | Elastic compute and storage turn infrastructure into a utility. | Scaled | Still expanding, now pulled forward by AI training and inference. |
| 2016-2021 | Blockchain | Open settlement networks create scarce digital assets and programmable money. | Evolved | Less a new internet, more a financial settlement and coordination layer. |
| 2021-2022 | NFTs | Tokenized media would reinvent art, games, identity and ownership. | Fizzled | Useful mechanics survived in tickets, loyalty, games and provenance; the JPEG asset class did not. |
| 2021-2024 | Metaverse | Persistent 3D worlds would become the next office, mall and social graph. | Fizzled | The broad social-platform story fizzled. Useful pieces survived in spatial computing, game engines, simulation and training data pipelines. |
| 2022-2026 | Generative AI | Foundation models make language, code, images, audio and video programmable. | Scaled | The current thing: a broad general-purpose platform, not just a consumer app cycle. |
| 2025-2026 | AI agents | Models stop only answering and begin planning, using tools and executing workflows. | Evolved | Likely evolves into workflow software, RPA replacement, coding systems and machine customers. |
| 2025-2030 | Robotics / physical AI | AI leaves the browser and acts through robots, factories, labs, vehicles and homes. | Frontier | The credible next big thing, if embodied data, hardware cost and reliability improve together. |
| 2025+ | Quantum | New hardware changes cryptography, chemistry, optimization and materials. | Frontier | Strategic frontier, not the current adoption wave. |
| 2025+ | Synthetic biology | Biology becomes easier to design, simulate, measure and manufacture. | Frontier | High-upside branch, likely slower and more domain-specific than software AI. |
The useful question is not "what is hyped?"
It is whether the wave has a route from custom novelty to product and then utility. Internet, mobile and cloud crossed that path. Blockchain partly did. NFTs mostly did not. AI is crossing it now.
Cloud open
Gartner forecast public cloud end-user spending at $723.4B in 2025, up from $595.7B in 2024.
Blockchain open
Speculative assets dominated attention, but stablecoins and custody rails survived.
NFTs open
The retail image-collectible market collapsed from peak attention into a smaller niche.
Metaverse open
Meta's consumer metaverse push collapsed as a platform thesis; enterprise and gaming niches remained.
Generative AI open
S&P Global counted over $200B of AI company funding in 2025, roughly half of technology funding.
AI agents open
McKinsey calls agentic AI one of the fastest-growing 2025 trends, though still early versus established trends.
Robotics / physical AI open
IFR reported 542,000 industrial robots installed in 2024, more than double the level ten years earlier.
Quantum open
McKinsey frames quantum as transformative in some domains, but still requiring major advances before broad business impact.
Synthetic biology open
AI helps protein, molecule and lab automation work, but production cycles remain physical and regulated.
Autonomous business, AI agents, embodied AI, humanoid robots, physical AI.
Agentic AI, semiconductors, quantum, AI as a cross-domain accelerator.
Cloud spending and hybrid-cloud forecast.
Industrial robot installation data and growth over ten years.
AI funding concentration and 2025 funding scale.