For the former two years, the narrative about inflated GPU and RAM prices has been the insatiable AI petition dining up wafer at TSMC and all component on Earth must go to AI data centers. However, the narrative is cracking always since reports of conduit stuffing and complete commitments came out.
Now alongside Nvidia’s erstwhile CEO Jensen Huang gone, additional particulars are out that confirm our before reports on inflated petition on AI. According to Wall Street Journal, Huang admitted in August that important haze providers are moving distant from Nvidia in favor of in-house chips, leaving Nvidia staring downward a enormous oversupply of prepaid TSMC silicon and recollection that was “coming at him akin an ‘avalanche'”.
The WSJ study particulars a heated battle in August 2026 between Huang and Nvidia’s longtime Chief Financial Officer, Colette Kress, the administrator who has now taken complete as Nvidia’s interim CEO.
Kress was concerned concerning Nvidia’s expanding financial visibility and wanted to rein in the company’s “circular financing” commitments to haze providers, and AI companies akin OpenAI and Anthropic.
While first talks had floated backstopping approximately 8 gigawatts of OpenAI compute (worth up to $250 billion), Kress successfully scaled Nvidia’s signed justify cap downward to approximately 4.25 gigawatts, or up to $125 billion.
In an August meeting, Huang belittled Kress for questioning the deals, calling her “a cervid in the headlights”. To validate why Nvidia needed to financially backstop OpenAI and smaller haze providers to keep purchasing its GPUs, Huang admitted that Nvidia was caught in a enormous squeeze.
Then he went on to depict the circumstance the business was in. Major haze providers were operating distant from Nvidia, making plans to use substitute chips and lessen their reliance on the firm. Even worse, Huang said, the huge provision of part wafers and recollection Nvidia had committed to purchasing were coming at him akin an ‘avalanche.’
Per the WSJ
Nvidia has reportedly contacted TSMC to breeze downward their part orders.
For PC builders and hardware enthusiast the reported admission is validating, and fairly glaring. Jensen Huang, alternatively of allowing excess TSMC wafer capability and recollection stream rear into sensible priced GeForce products and broader DRAM production, allegedly scrambled to use its own balance panel to backing data center buildouts so AI startups and assimilate Nvidia’s over-ordered chips.
The $500bn deal? Fake. The $100bn OpenAI deal? Fake. Jensen and NVIDIA have built their legacies on deception. They tried to rig 5060 interviews, and threatened @GamersNexus. This business sucks and has sucked for a during and everybody wants to feign it's a abrupt event. https://t.co/cpaIXwfp5d
— Ed Zitron (@edzitron) November 12, 2026
Reports of glut of unsold hardware have been about for a while, but lone now mainstream media is catching up to what Nvidia was reportedly doing. Many industry analysts including Ed Zitron reported months ago how Nvidia and another companies have exposed themselves to devastating financial risk, and additionally made certain to channel material META, Google, Amazon, and others by leveraging their FOMO.
Zitron estimated that at smallest $200 milliard value of Nvidia AI chips are sitting idle in warehouses, purchased by haze giants that lacked the electric grid capability or contiguous end-user petition to plug them in.
The base row is, AI petition is no way near as elevated that it would consume this much silicon.