"Foothills of the Singularity": What Google I/O 2026 Really Told Us About Where AI Is Heading.
Demis Hassabis ended Google I/O 2026 with a line about the singularity. A breakdown of what Google actually shipped — Gemini 3.5 Flash, Gemini for Science, an AI-powered cursor — and what the framing tells us about where AI is going.
Introduction
Every year, Google I/O serves as a barometer for where the tech giant believes AI is going. But this year’s conference felt different. Between a flashy parade of new models, an AI-powered cursor, a scientific research platform, and a closing statement from DeepMind CEO Demis Hassabis invoking the singularity, Google sent a clear message: the era of incremental AI features is over. We are now in something bigger.
”We’re at the Foothills of the Singularity”
The most talked-about moment from Google I/O wasn’t a product demo — it was a philosophical declaration.
After more than two hours of announcements that included flashy new models, AI agents, a smarter search box, tools for work and video creation, and smart glasses, Google DeepMind CEO Demis Hassabis delivered a poetic, if slightly abrupt kicker: “We’re at the foothills of the singularity,” he said on stage in Mountain View.
The decision to end the show that way was a deliberate one. “We debated it back and forth,” Hassabis said in an interview. “I was closing, and I wanted to be authentic about what I’m thinking with AGI. The singularity, at least my interpretation of that word and that term, means the era that we’re in.”
For Hassabis, the singularity isn’t a science fiction moment — it’s a description of compounding capability.
Google showed off how its Antigravity 2.0 product can autonomously build a computer operating system for less than $1,000 — a task that would have required teams of software engineers months to complete in the pre-AI era. Hassabis sees the increase in machine autonomy as one of the key steps on the gradual march toward the singularity. “This year, I really felt … that it’s the beginning. Agents are starting to work, becoming useful harnesses.”
Gemini 3.5 Flash and the Enterprise Price War
On the product front, Google’s headline model announcement was Gemini 3.5 Flash.
Google launched Gemini 3.5 Flash and cut the Ultra subscription from $250 to $200 per month. The pricing cut targets enterprise clients that OpenAI and Anthropic have dominated so far.
Gemini 3.5 Flash is now generally available — frontier-level intelligence at 4x the speed of comparable models, priced at $1.50/$9 per 1M tokens, with a 1M context window, achieving 76.2% on Terminal-Bench 2.1 and beating Gemini 3.1 Pro on coding and agents.
The aggressive pricing signals that Google is playing a long game: accept margin compression now to win the developer ecosystem before it entrenches on competitors.
Gemini for Science: A Platform for Discovery
The most substantive announcement for the research community was the unveiling of Gemini for Science.
The big scientific announcement at I/O was the new Gemini for Science package, which unites several of the company’s LLM-based scientific systems under one brand. This includes the hypothesis-generating AI Co-Scientist and algorithm-optimizing AlphaEvolve, which are still not publicly available — but as Google is now allowing any researcher to apply for access to Gemini for Science, they may soon see wider adoption in the scientific community.
The early praise has been extraordinary.
Scientists who were involved in early testing are enthusiastic about their potential: Gary Peltz, a Stanford geneticist, compared using the AI Co-Scientist to “consulting the oracle of Delphi” in a Nature Medicine article.
Across the industry, agentic researcher systems are showing real potential. This week, OpenAI announced that one of their models had disproved an important mathematics conjecture — perhaps the most meaningful contribution that generative AI has made to mathematics so far, according to some mathematicians.
The AI-Powered Cursor: Rethinking Human-Computer Interaction
One of the more quietly radical announcements was a reinvention of something most users never think about: the mouse cursor.
Google DeepMind announced a research effort to transform the standard computer mouse cursor into a context-aware, AI-powered tool, marking what the company described as the first major rethinking of the cursor in more than 50 years. The project integrated Google’s Gemini AI model with an experimental context-aware mouse pointer, allowing the system to understand where a user clicks, what they are clicking on, and the likely intent behind the interaction.
A feature called Magic Pointer will soon roll out on the forthcoming Googlebook laptop platform. The technology will also allow users of Gemini in Chrome to point at specific parts of a webpage and ask questions, rather than composing a full text prompt.
It sounds incremental, but the implications are significant — AI that meets you inside your existing workflow rather than demanding you switch contexts.
DeepMind, AlphaFold, and the Next Frontier
DeepMind worked separately and autonomously within Google for years, but the AI race has thrust the division into a central product role at the company. Now, DeepMind’s work is touching nearly everything the company makes — even Google Maps now uses the Genie world model to enhance the Street View feature.
After the AlphaFold Nobel, the company spun off Isomorphic Labs, which is not about “any one particular drug or one particular disease” but aims to create a technology that cures “hundreds of diseases.”
It’s not surprising that Google is assigning its best minds to the coding problem, as the company has recently taken a reputational hit because its coding tools don’t currently stand up to those offered by Anthropic and OpenAI. But it may also signal a prioritization of agentic science on Google’s part, as coding abilities are key to the success of some of those systems.
Opinion: The Singularity Framing Is Deliberate — and Important
Demis Hassabis is not given to hyperbole. When he invokes the singularity — however cautiously — it’s worth taking seriously. What’s striking about Google I/O 2026 is not any single product announcement, but the coherence of the vision: AI embedded everywhere, at every level of the stack, from the cursor on your screen to the hypotheses your lab generates.
Hassabis reminded everyone that the flurry of Google’s announcements were all linked by a common technology. The narratives around Google falling behind competitors like OpenAI and Anthropic — and then zooming ahead, and falling behind again — are ever-changing because the technology is fluid. We’re still very early in the AI race.
The foothills metaphor is apt. The mountain is very tall. And the climb has barely begun.