Cerebras CEO Andrew Feldman to address AI scaling at Disrupt 2026
Cerebras Systems CEO and co-founder Andrew Feldman will take the stage at TechCrunch Disrupt 2026 to examine whether artificial intelligence can continue scaling as demand for compute, energy and infrastructure rises. The session, titled Can AI Keep Scaling?, is scheduled for October 13–15 at Moscone West in San Francisco.
The appearance comes as Cerebras expands its wafer-scale computing strategy. The company raised $5.5 billion in its May initial public offering and signed a multiyear agreement with OpenAI to deploy 750 megawatts of Cerebras systems between 2026 and 2028.
Wafer-scale computing as an alternative
Founded in 2015, Cerebras was built around an approach that departs from conventional AI chip design. Instead of cutting a silicon wafer into individual chips, the company develops a processor built on the wafer itself for demanding AI workloads.
Feldman co-founded Cerebras after leading the energy-efficient microserver company SeaMicro, which AMD acquired in 2012. He also held leadership roles at Force10 Networks and Riverstone Networks. At Disrupt, he is set to discuss what may happen if current AI hardware reaches its limits and how Cerebras is approaching those constraints.
The programme also places the discussion alongside AI infrastructure scaling at TechCrunch Disrupt 2026 in a broader examination of how companies are preparing infrastructure for increasingly demanding AI workloads.
Compute growth depends on physical capacity
Cerebras introduced its CS-4, the latest generation of its wafer-scale AI infrastructure, in August. The company also reported more than 600 megawatts of data-centre capacity either live or under contract for delivery by the end of 2027.
Its plans extend beyond processors. Cerebras said it was increasing manufacturing capacity by more than tenfold during 2026, planned to bring its first European data-centre capacity online this year, and aimed to expand European capacity to 200 megawatts by the end of 2027.
Those figures underline the scope of the scaling challenge: deploying AI compute requires data centres, electricity, cooling and manufacturing capacity as well as more capable processors. TechCrunch expects more than 10,000 founders, investors, operators and technology leaders at the event, with more than 200 sessions, 250-plus speakers and 300 exhibiting startups.
What businesses should consider
For organisations building, funding or deploying AI, the session frames infrastructure as an operational dependency rather than a purely technical specification. Capacity plans should account for the availability of power, cooling, data-centre space and hardware supply alongside the performance of the processors selected.