The MegaETH AI pivot has arrived with a restructuring attached. MEGA, the organization formerly known as the MegaETH Foundation, announced a 20% workforce reduction on November 13 as it retools the Ethereum scaling project around AI products, with co-founder Shuyao Kong confirming the cuts and a $30 million treasury commitment to back the new direction.
Departed employees reportedly came from the blockchain, wallet, ecosystem-support and communications teams, though the company described the reduction as the result of a team reorganization rather than attributing it solely to the AI move. Exact headcount was not disclosed.
From layer-2 to the inference layer
MEGA’s plan is a public AI inference market where users can call competing AI service providers through the network, targeting a December 2026 launch. The proposal comes via a whitepaper from co-founders Yilong Li and Lei Yang, and the $30 million earmark is meant to seed initial products, with AI services first in line.
It is a recognizable pattern by now: crypto networks chasing AI workloads as the market prizes anything with inference, agents and usage in the same sentence. MegaETH built its reputation on speed as an Ethereum layer-2, and high-frequency inference traffic is a plausible neighbor for an infrastructure stack designed for low latency.
The stakes of the MegaETH AI pivot
Layoffs at a well-funded scaling project in the middle of a sixth straight weekly crypto loss are not exactly shocking, but the strategic shift is meaningful. Layer-2 chains compete on fees and ecosystems that are hard to win outright; an open inference market competes on a narrative the entire industry is currently underwriting.
The risks are plain. December is a demanding deadline for a public market, AI inference economics are unforgiving, and the teams cut were the ones shipping the old roadmap. Still, MEGA is betting that the next chart that matters won’t be transactions per second but tokens served per second, and it is willing to reorganize the company to make that bet.
