AI layoffs came for MEGA three days after the company told the world it was spending $30 million on AI, and that order of events is the entire story. Co-founder Shuyao Kong announced a 20% staff reduction on November 13, 2026, days after the organization formerly known as the MegaETH Foundation committed treasury money to first-party products that begin with an AI inference marketplace (Crypto.News). The people who lost their jobs were not the ones who needed to be convinced the pivot was smart. They were the ones who built the chain, the wallet and the developer support that was supposed to make it worth using. A company can change direction in a memo. The people standing where the old direction pointed do not get a vote.
The disclosure was thin on the details that matter to a payroll. Kong did not say how many people were affected or which teams lost the most people, describing the change only as new capabilities brought in alongside a reorganized staff. What she did confirm is that the departing employees had been there since the early days, working on blockchain development, wallet operations, ecosystem support and communications (Crypto.News). Founders, early employees and the core engineering group stayed. Read that list twice: the outward-facing functions — the ones whose job was to attract developers and projects to the network — were the functions the company decided it no longer needed at full strength.
Shuyao Kong — The People Who Built the Onboarding Pipeline Were the First Thing Cut
Kong framed the reduction as a team that had spent months preparing to move in a new direction, and she credited the people leaving as the ones who helped build the organization from its earliest days. That is a gracious sentence and a revealing one. If the people who built the ecosystem arm were let go, the ecosystem arm was the thing being written down — not trimmed, not re-scoped, but judged against a future that no longer includes it. Core engineering survived because the next product needs engineers. Ecosystem support did not survive because the next product does not need developers arriving from outside.
That is a real strategic argument, and it is also the argument every company makes right before it discovers it has cut the only people who knew why the last strategy failed. Kong told The Block that most of the applications the network’s support program had backed were no longer being built on MegaETH at all, and that the network had supplied technical help, management help and market-making without taking equity, governance rights or any ownership in those projects (Crypto.News). Fund other people’s apps, own none of the upside, watch them leave. That is not a mystery that needed three years to solve. It is a business model with the profit line removed.
MEGA’s Treasury — AI Layoffs Landed Three Days After the $30M Pledge
The $30 million is not new money raised from investors. It is a reallocation of MEGA’s existing treasury toward products the organization builds and owns itself, part of a rebrand that dropped the word “foundation” from the name because a foundation supports a network and a product company sells things. The first thing it intends to sell is a public marketplace for AI inference capacity, targeted for December 2026, where buyers and sellers of compute meet in one venue and MegaETH’s cryptography is supposed to make the machinery underneath verifiable (Bitcoin Foundation).
The organizing phrase for all of this is “purified intelligence,” which Kong has used to describe a belief that cryptography and distributed systems will change how the world produces intelligence. Strip the branding and the pitch is straightforward: centralized AI providers ask customers to trust them about which model answered, at what price, on whose hardware. A verifiable marketplace replaces that trust with a proof. It is a genuinely interesting idea. It is also a crowded room full of well-funded companies chasing the same unmet demand, and MEGA is entering it with a December deadline, a shrunken team and a treasury it just spent on itself.
MegaETH Deposits — $245 Million Became $12 Million and the Staff Shrank Anyway
The financial backdrop explains the urgency better than any statement. MegaETH’s total value locked peaked above $245 million in May 2026 and had fallen to roughly $12 million by the time the restructuring landed — a decline of about 95% (Bitcoin Foundation). Ninety-five percent is not a bad quarter. It is a verdict from the only constituency that cannot be reorganized: the people who decide whether to leave their money in a network. When deposits leave, the fees leave, and when the fees leave, the only remaining lever is the payroll.
Ecosystem teams are the easiest line to cut because their output is measured in a metric that had already stopped moving. Cutting them reads as discipline in a board deck. In practice it removes the sales force for a network that still needs developers, and it signals to every remaining project on MegaETH that the organization is now primarily interested in its own products. A chain whose own builder stops recruiting outside builders is a chain announcing what it thinks of its own market (Crypto Briefing).
Axel’s Take — A Pledge Is a Press Release, a Layoff Is a Receipt
Here is what separates this from the usual AI restructuring story. MEGA did not cut staff to pay for AI it had already bought from someone else, the way a media company cuts writers to license a model. It cut staff to fund AI it intends to sell, using money it already held, three days after announcing the ambition. The sequence says something honest about the economics of the AI build-out: the capital has to come from somewhere inside the building, and the somewhere inside the building is almost always the people. Nobody runs a treasury reallocation through a hiring freeze. The $30 million had a funding source, and the funding source had names.
December is now the whole test. Kong has said the marketplace’s operating details, its providers and its available models are being held until launch. If the inference market ships on schedule and finds paying customers, the layoffs look like a hard but coherent bet and the people who left become a footnote in a turnaround story. If it slips, MEGA will have spent its treasury, shed the team that made it a network, and bought a launch date it could not hit. Both outcomes were decided in the same week, by the same people, and only one of them gets written about next year.
There is a person at the center of this who will never appear in a press release. Someone spent three years building wallet tooling for a chain that was going to be the fast one, the one with real throughput and real users, and then sat in a room where the words “new capabilities” were used to describe their departure. They are not a line item in a treasury reallocation, and they are not a rounding error in a 95% decline. They are the reason the product shipping in December has anything to ship on top of. Companies do not pivot. People get pivoted around, and the ones doing the pivoting rarely have to explain it to the ones being moved. The chain will survive this. It usually does. The careers are the part nobody publishes numbers about.
Sources: Crypto.News, Bitcoin Foundation, Crypto Briefing.