Nvidia currently finds itself in the spotlight for several reasons at once: according to a report by Crypto Briefing, Apple has overtaken Nvidia as the world's most valuable company, which market observers attribute to a broader shift in the AI market. At the same time, Nvidia CEO Jensen Huang has personally spoken out, according to TradingKey, to address doubts about the stock's future performance in the second half of 2026. In parallel, new technologies and financing models around Nvidia chips continue to fuel the debate over the future of AI infrastructure.
- Apple has overtaken Nvidia as the most valuable company, according to Crypto Briefing – a sign of a shift in the AI market.
- Jensen Huang is personally responding to market doubts about Nvidia's stock for the second half of 2026, according to TradingKey.
- The new Vera Rubin architecture lowers the cost of post-training AI models through a seven-chip codesign, according to Tech Times.
- A $400 million deal shows that inference chips are increasingly replacing Nvidia GPUs as loan collateral, according to Tech Times.
- Nvidia is expanding its Physical AI offering in Japan with Cosmos 3 Edge and new industry partnerships, according to Pulse 2.0.
Why did Apple overtake Nvidia as the most valuable company?
According to Crypto Briefing, Apple has overtaken Nvidia as the most valuable company amid a shift in the AI market. The report frames this as a reflection of changing market dynamics in the AI sector, without providing further details on the exact background.
How is Jensen Huang responding to doubts about Nvidia's stock?
According to TradingKey, Jensen Huang has personally addressed the market doubts that have emerged amid recent price movements. At the center is the question of whether Nvidia's stock can continue to rise in the second half of 2026 – a topic that, according to The Motley Fool, is also occupying other analysts who are highlighting alternative AI chip stocks amid a current sell-off.
What does Nvidia's new Vera Rubin architecture bring?
According to Tech Times, the Vera Rubin platform lowers the cost of post-training token handling for AI models through a seven-chip codesign – a key factor for the efficiency of future AI training infrastructure.
How is the role of Nvidia GPUs as loan collateral changing?
According to Tech Times, inference chips are increasingly replacing traditional Nvidia GPUs as collateral for loan financing in the AI infrastructure market, as part of a deal described as landmark by General Compute worth $400 million.
The General Compute deal is considered a landmark transaction, according to Tech Times, because for the first time inference chips rather than traditional Nvidia GPUs are serving as collateral for financing of this scale. This suggests that financing models in the AI infrastructure market are diversifying, even as Nvidia hardware continues to play a central role.
On the product side, Nvidia also remains active: PC Guide reports on a $400 discount on an RTX 5070 system at Newegg, while quasa.io highlights the Metropolis 3.2 and DeepStream 9.1 updates, which the report says enable up to six times faster development of vision AI agents. Additionally, Nvidia is expanding its Physical AI offering in Japan with the Cosmos 3 Edge platform and new industry partnerships, according to Pulse 2.0, while MarkTechPost has published a tutorial on fine-tuning Qwen3 with Nvidia NeMo AutoModel based on LoRA.
