AI infrastructure is becoming an energy story as much as a software story. At TechCrunch Disrupt 2026, that tension will move into focus on the Smart Systems Stage, where speakers will examine the power, grid, and infrastructure challenges behind the next wave of technology growth.
The event runs from October 13-15 at San Francisco's Moscone Center. Its Smart Systems Stage agenda connects several issues that are often discussed separately: commercial fusion, aging grid infrastructure, data center electricity needs, and the strain AI growth is putting on the broader economy.
AI needs more than code
The source article frames the core issue clearly: AI does not run on code alone. It depends on large amounts of power, and that demand is increasingly described as a bottleneck.
That matters because the conversation around AI infrastructure often starts with models, chips, and data centers. The Smart Systems Stage shifts attention to the physical systems that must support those tools. Without enough electricity and modern infrastructure, compute growth can become constrained by the power system around it.
TechCrunch says the stage will bring together energy, infrastructure, and technology. The planned sessions are built around a practical question: what does it actually take to power the next decade of innovation?
Fusion gets a grid-focused discussion
One major session, Bringing Fusion to the Grid, will feature leaders from Commonwealth Fusion Systems and Helion. The discussion is set to cover the breakthroughs moving commercial fusion forward, the challenges that remain, and what would be required to bring fusion power onto the grid at scale.
The listed speakers are David Kirtley, CEO, Helion, and Brandon Sorbom, Chief Science Officer, Commonwealth Fusion Systems. Their session places fusion in a practical infrastructure context rather than treating it only as a scientific milestone.
That framing is important within the limits of the agenda itself. The issue is not simply whether fusion advances, but how any future power source connects to the grid and supports real demand. For AI infrastructure, the question of scale is central because the article presents electricity as a possible constraint on growth.
Startup operators meet hard engineering questions
The agenda also includes How Twilio's Founder Is Tackling the Power Problem, a fireside chat with Inertia CEO Jeff Lawson. The source describes Lawson's path from founding Twilio to leading Inertia, one of the best-funded fusion power startups in the world.
The session is expected to connect company-building experience with energy infrastructure work. Specifically, it will address how scaling Twilio is shaping Lawson's approach to talent, timelines, and difficult engineering questions at Inertia.
This part of the program underlines a broader theme of the Smart Systems Stage: the power problem is not only technical. It also involves hiring, execution, investment, and long development timelines. Those are familiar challenges for founders and operators, but the energy sector gives them a different level of complexity.
The grid becomes a startup issue
Another session, Rewiring the Grid for the Electric Age, focuses on the infrastructure that already carries electricity. TechCrunch says electricity demand is growing faster than the systems built to support it.
The panel will examine what it takes to modernize the power system, where investment is flowing, and how utilities, startups, and technology providers are working toward a more resilient, flexible grid.
The announced speakers include Drew Baglino, Founder & CEO, Heron Power, and Apoorv Bhargava, CEO and Co-founder, WeaveGrid. More speakers are still to be announced.
For founders and operators, this topic sits close to the center of the AI infrastructure debate. If the grid cannot keep pace with demand, then new software products and compute-heavy services may depend on infrastructure decisions outside the software industry itself.
Data centers face the electricity race
The session AI's Power Problem brings the discussion directly back to compute demand. According to the source, data center operators and energy companies are racing to secure power and expand infrastructure before electricity becomes the bottleneck that slows AI's next wave.
The listed speakers are Sara Spangelo, President & Co-Founder, Ambrosia Energy, and Bill Thayer, SVP, Head of Datacenter Solutions, Bloom Energy. More speakers are also expected for this session.
This discussion connects AI growth to the operational realities behind it. Data centers need access to power, and the article presents that access as something companies are actively trying to secure. The implication is straightforward: AI infrastructure planning cannot stop at servers and software.
Across the Smart Systems Stage lineup, TechCrunch is positioning power as a strategic constraint for the startup ecosystem. The stage is aimed at people building energy breakthroughs, rethinking grid infrastructure, or trying to understand what is really constraining AI's growth.
Disrupt 2026 attendees will also have access to other stages, networking, side events, and the broader speaker lineup. The current pricing window is closing, and TechCrunch says attendees can save up to $330 today.