I won’t engage in a discussion regarding the US/Israel political goals, as I am not a geopolitical expert, and even experts have little explanation other than Trump having been fooled by Netanyahu into this mess, but oil exporters told OPEC+ that even when the Strait of Hormuz reopens, supply disruptions resulting from the prolonged closure of the Strait will persist.

Price-wise, though, I would say the latter would mean lower global prices and potentially some pressure on oil companies’ earnings. The most important is the reopening of the Strait.

This morning, US crude is slightly lower, hovering around $92pb. Short-term risks remain two-sided. It is possible that the barrel of US crude rebounds past the $100pb mark if peace negotiations stall. It is possible that we see prices pull back toward $80pb if the Middle East fighting ends.

Yet, given the rapidly falling global oil reserves, the next leg up threatens to push oil higher than the $120pb mark. Above this level, demand destruction would likely keep prices capped. Unless the war extends into July, I wouldn’t yet expect prices to rise to the $150–160pb levels suggested by some industry leaders.

Looking at oil stocks, they performed well yesterday on the back of the rebound in oil prices. Yet with the initial impact of higher prices already priced in, and in light of a prolonged period of supply disruptions, we could see SPDR’s Energy ETF step into the medium term bearish consolidation zone below the $55 level/100-DMA.

Next big thing

Zooming out, the S&P500 rose to a fresh ATH yesterday, led higher by tech – of course. Nvidia jumped more than 6% yesterday after unveiling a new superchip for PCs (applause). Jensen Huang said it was the ‘first big reinvention’ in 40 years. It aims to allow large AI models to run directly on a Windows computer instead of sending requests to a data center and getting them back.

In other words, Nvidia wants to turn every computer into an AI machine, capable of handling complex tasks across your programs and files. This is fantastic for AI adoption: it means faster responses, lower cloud costs and greater privacy – think of sensitive data that no longer needs to leave your device. For medical and legal industries, the latter could be a game changer.

And for data centers, it wouldn’t be a worry – quite the contrary. Companies would face fewer capacity constraints. Demand will remain robust because the most advanced AI models will still require enormous computing power to train and run. If anything, putting AI on every desktop could accelerate adoption and increase demand for the cloud infrastructure behind it.

So Nvidia opened a fresh revenue avenue for itself yesterday, justifying investors’ enthusiasm. HP, Dell and Lenovo gained between 6% and 10%, extending an almost vertical ascent over the past few months, while AMD and Intel fell as competition in PC chips intensified. Microsoft advanced more than 2% after unveiling the first Windows PC to run on an Nvidia main processor.

Interestingly, the news only helped TSMC gain 1% in Taiwan and couldn’t help the Kospi extend gains. Investors preferred to pocket some profits against the backdrop of a worsening geopolitical outlook.

A finance story

Another major story was Anthropic submitting draft paperwork for a fall IPO, aiming to leapfrog OpenAI – which is also expected to file for an IPO in the coming days. The valuations are huge for these ultra-promising but cash-burning companies.

What's perhaps more interesting is that the AI race is no longer being funded solely by venture capitalists willing to lose money for a decade in exchange for a shot at changing the world. The financing is becoming increasingly institutionalized.

Just yesterday, Alphabet announced plans to raise $80 billion to fund its AI ambitions – one of the biggest stock deals in history – including a $10 billion investment from Berkshire Hathaway. This means that AI is increasingly becoming a financing story as well. And the deeper traditional finance gets involved, the more the AI story shifts from a technology narrative toward a financing and credit narrative.

What’s the risk? A failure of OpenAI or Anthropic – God forbid – would not likely trigger a systemic financial event. But the growing web of equity investments, debt financing, private credit facilities, infrastructure spending and long-term purchase commitments means that AI-related losses are increasingly finding their way into pension funds, insurers, asset managers, corporate balance sheets and the broader economy. We’re all in the same boat.

Of course, S&P500 earnings are expected to accelerate by 24% this year on the back of AI. Goldman Sachs’ single-stock one-month put-call skew fell to the lowest level on record, meaning investors have become extraordinarily optimistic. No more bears. Just bulls.

Two charts from Axios warn that corporate profits now make up around 12% of gross domestic income – the highest level since the 1950s – while compensation going to employees has fallen to 51%, pointing to a two-speed income growth story for US households: those who invest versus those who do not.

So to get ahead, one increasingly needs to invest. But if everyone is already in, are valuations being pushed beyond what fundamentals can justify?

AI will survive, just as the internet survived the dot-com bubble. The question is whether the AI rally can avoid a road accident, or whether it will eventually resemble the dot-com bust of the early 2000s.

No one has the answer.