Personalities give character to the otherwise dry realm of science and technology. How plain do Newton’s laws of motion sound without the his name? Or Heisenberg’s uncertainty principle? These big personalities tend to have massive egos as well, often leading to intense competition and disputes. Newton and Gottfried Leibniz fought over who invented calculus. Niels Bohr and Albert Einstein waged a series of intellectual battles over the interpretation of quantum mechanics. And George Westinghouse and Nikola Tesla challenged Thomas Edison over electric power transmission systems—alternating current vs direct current.
Today, that fire has moved from the laboratory to the silicon foundry. We are living in a time where names like Altman, Huang and Nadella are becoming as synonymous with the 21st century as Edison and Tesla were with the 20th. But, in this war, the battle lines are more complex.
The many battles in the war
The AI war is being fought on three distinct fronts. First, there are the Large Language Models (LLM). For a while, OpenAI sat unchallenged on the throne with its ChatGPT. However, the landscape has since shifted into a brutal war of attrition. While OpenAI and Anthropic remain the ‘intellectual’ leaders, Google has successfully weaponised its ecosystem, integrating Gemini into every corner. A surge of ‘open-weight’ models—led by Meta’s Llama and China’s DeepSeek—have commoditised AI. In the LLM space, the ‘winner’ is no longer the one with the smartest model, but the one with the most efficient distribution. At the moment, Gemini seems to be winning the war.
The second front is the chipmakers, and here, the spoils of war are far more concentrated. NVIDIA remains the undisputed leader, effectively acting as the ‘arms dealer’ to everyone. However, the tide is turning toward custom silicon. Amazon, Google and even Tesla are desperately trying to nibble into NVIDIA’s territory to reduce their dependency on a single source of power.
Finally, we have the gadget makers. This is a chaotic front. Traditional smartphone manufacturers are locked in a race to prove their devices are ‘AI-native’, while a host of startups are attempting to kill the smartphone altogether with wearable ‘companions’.
The everyday AI
Where is this war actually being won? By 2026, standalone AI platforms have surpassed 1.5 billion monthly active users. However, the vast majority of people don’t go to AI; it comes to them.
For instance, Microsoft 365 and Google Workspace have turned AI into a silent ghostwriter for billions of emails and documents. Meta AI, integrated into WhatsApp and Instagram, has become the default ‘search’ for the younger generation, particularly in emerging markets like India and Brazil. Then there is the invisible layer, where AI optimises traffic lights, manages power grids or filters your spam—tasks where the users do not even know they are interacting with a machine.
Perhaps the greatest casualty of the AI war is the traditional search engine. The ‘blue links’ that defined the internet for two decades are effectively dead. In 2026, AI overviews are the default.
Search has shifted from ‘discovery’ to ‘synthesis’. Instead of navigating a website to find an answer, the AI reads the web for you and presents a summary. This has triggered a seismic shift in SEO (search engine optimisation) to GEO (generative engine optimisation). Businesses no longer fight to be ranked top on a page; they fight to be the ‘cited source’ inside an answer. If you aren’t referenced by the LLM, you don’t exist.
New rules
While smartphones still dominate AI usage—accounting for over 70 per cent of consumer AI interactions—the era has triggered the invention of entirely new device categories.
Devices like the Limitless Pendant and Plaud NotePin act as passive observers, recording and summarising our daily lives so we never ‘forget’ a conversation. We’ve also seen Spatial AI gadgets, such as smart glasses that actually understand what the wearer is looking at, providing real-time translation or step-by-step repair instructions for a leaky faucet.
In the corporate world, 2026 is the year of the ‘Great ROI Appraisal’. The winners: Companies that have moved past ‘chatbots’ and integrated AI into their core unit economics. Logistics firms using AI for real-time route optimisation and coding houses using AI to double their output are seeing massive gains.
The losers: AI-washing firms that added a thin layer of GPT to their existing products without changing their business model. Investors are becoming increasingly discerning, punishing companies that cannot show how AI translates into actual profit.
The next frontier
The conversation has moved beyond models that ‘talk’ to models that ‘do’. Agentic AI—systems capable of planning, using tools and executing multi-step tasks autonomously—is where the industry is headed.
In 2026, an AI agent doesn’t just help you write an email about a flight delay; it logs into your airline account, finds a new flight, books a seat and updates your calendar. This shift is turning AI from a novelty into a utility, but it also raises the stakes for security.
Is it a bubble?
A bubble, by definition, is a good thing that has gone too far. With trillions of dollars flowing into AI, the question of a ‘bubble’ looms large. Critics point to a worrying trend: circular deals.
Many AI startups receive billions in funding from Big Tech giants, only to immediately spend those same billions back on those same giants’ cloud credits and chips. This creates artificial revenue that can mask a lack of true market demand. While some analysts compare this to the 1999 dot-com crash, others argue the infrastructure being built (data centres and energy grids) has real, tangible value that the 1990s fibre-optic ‘dark cables’ did not.
Whether a bubble or a boom, the sheer scale of investment—projected to hit $4 trillion in by 2030—means the AI war will not end quietly. And the winners will be the ones who not only invent the technology but also claim its name. We may not remember the specific version of an LLM, but we will remember the personalities who dared to automate our lives.