When OpenAI quietly signaled last week that it was “saying goodbye” to its video-generation app Sora, the brevity of the announcement masked a deeper reality: one of the most hyped products in generative AI had become too expensive to sustain.

Behind the decision lies a hard constraint now shaping the industry — compute. Running Sora, by multiple estimates, was costing the ChatGPT maker close to $1 million a day, a figure that appears to have tipped the internal balance against the product as the company sharpens its focus ahead of a potential public listing.

Unlike text-based systems such as ChatGPT, video generation operates at the extreme end of compute intensity. Each prompt requires the model to generate sequences of frames, maintain temporal consistency, simulate motion physics, and, in many cases, synchronize audio. The result is an exponential increase in GPU usage per query.

At scale, that becomes unsustainable.

Sora’s early trajectory suggested a breakout success. Within weeks of its September debut, it climbed to the top of app store rankings, amassing millions of downloads and dominating online discourse. Yet the initial surge proved difficult to maintain. By early 2026, download momentum had slowed sharply, even as the cost of serving each user session remained high.

That mismatch, declining marginal growth against persistently high operating costs, appears to have been decisive.

Internally, the calculation is increasingly about compute allocation efficiency. Every GPU cycle spent rendering video is a cycle not used to train or serve higher-margin products. For a company competing at the frontier of AI, where training next-generation models can cost billions of dollars, such trade-offs are no longer theoretical. They have become existential.

OpenAI’s subsequent exclamation points in that direction. The company said it would redirect the Sora team toward world simulation research, a domain tied to robotics and embodied AI. That shift underlines reprioritization: moving away from consumer-facing novelty applications toward foundational systems with clearer long-term commercial pathways.

The decision comes at a crucial time in the ChatGPT maker’s history. As OpenAI edges closer to a possible IPO, investors are likely to scrutinize not just growth metrics but cost discipline. High-burn, low-monetization products such as Sora complicate that narrative. In contrast, enterprise-facing tools, coding assistants, workflow agents, and API services offer more predictable revenue streams and better alignment with compute spending.

The company’s evolving product decisions reinforce that shift. Features such as instant checkout and more experimental consumer-facing modes have been scaled back, while development has intensified around integrated “superapp”-style functionality designed for workplace productivity.

In that sense, Sora’s shutdown is less an isolated move than part of a broader restructuring of priorities.

There is also a competitive layer. Rivals are increasingly focusing on enterprise deployment, where reliability, latency, and cost-per-query matter more than viral appeal. In that environment, a product that consumes vast compute resources without a commensurate revenue model becomes difficult to justify.

Sora faced pressure on another front as well: governance. Video generation tools sit at the middle of ongoing concerns around deepfakes, intellectual property, and misinformation. Efforts to impose safeguards tend to increase operational complexity and, in some cases, reduce user engagement — further weakening the business case.

The convergence of these factors left Sora exposed.

The underlying technology, however, is unlikely to disappear. By shifting resources into world simulation, OpenAI is effectively repositioning video generation as an enabling layer for robotics and physical AI systems rather than a standalone consumer product. Models capable of simulating environments, motion, and object interaction are critical to training machines that can operate in the real world.

That reframing suggests Sora’s demise is not about technical failure, but about economic prioritization.

The decision highlights a defining feature of the current AI cycle: compute has become the industry’s scarcest resource. Companies are now forced to make explicit choices about where to deploy it, often at the expense of high-profile products.

In Sora’s case, the conclusion appears to be that the cost of running the platform, at scale, and without a sufficiently strong revenue engine, became too high to justify, even for one of the best-funded players in the field.

Gold prices edged higher on Tuesday morning but failed to alter what is shaping up to be the metal’s sharpest monthly decline since the depths of the 2008 global financial crisis.

The U.S.-Iran conflict, surging oil prices, and a stronger dollar continue to drive heavy liquidation across the bullion market.

Spot gold rose about 1 per cent in early trading to around $4,553.69 per ounce, while front-month futures climbed 0.6 per cent to roughly the same level. Even so, bullion remains firmly on course for its steepest monthly drop in nearly 17 years. Recent market data published by CNBC show gold is down between 11.8 per cent and 14.6 per cent for March, depending on the close, which would make it the worst monthly performance since October 2008.

The rebound comes amid continuing uncertainty over the war in the Middle East, now in its fifth week.

A report by The Wall Street Journal said President Donald Trump told aides he was prepared to halt military hostilities against Iran even if the Strait of Hormuz remained largely closed, a signal that briefly improved sentiment across markets.

Trump later wrote on Truth Social that Washington was “in serious discussions” with Iranian officials, while warning that if a deal was not reached quickly, U.S. forces would target electricity plants, oil wells, and the strategically critical Kharg Island.

The mixed messaging has left markets caught between hopes of de-escalation and the risk of a broader energy shock. That uncertainty was reinforced by Secretary of State Marco Rubio, who said in an interview that Washington’s objectives in Iran would take “weeks, not months” to achieve.

Reports that 2,500 U.S. Marines from the 82nd Airborne Division had arrived in the region over the weekend further underscored the risk that the conflict may yet intensify. Ordinarily, such geopolitical turmoil would strengthen gold’s appeal as a haven asset. Instead, the market has moved in the opposite direction.

The key reason is that the conflict’s economic transmission has come through energy markets rather than direct flight-to-safety buying.

The effective closure of the Strait of Hormuz, a major conduit for global crude and gas flows, has sent oil prices nearly 50 per cent above pre-war levels, sharply lifting inflation expectations. That has forced investors to reprice the outlook for U.S. monetary policy.

Markets that had earlier expected multiple rate cuts this year are now leaning toward one or none, a shift that has pushed Treasury yields and the U.S. dollar higher, both of which traditionally weigh on non-yielding assets such as gold.

Wayne Nutland, investment manager at Shackleton Advisers, said the market has reverted to its pre-Ukraine-war behavior.

“Prior to the Ukraine war, the gold price tended to be inversely correlated to real bond yields and the US dollar, with the gold price rising when those metrics fell, and gold falling when those metrics rose,” he said.

“The period after the Ukraine war upended these relationships, in particular in 2025 and into early 2026 when gold rose very strongly, far in excess of the moves suggested by those historic relationships.”

He added that the Iran war has now restored those traditional correlations.

“Bond yields and the U.S. dollar have both moved higher, and against this backdrop gold has demonstrated its traditional inverse sensitivity to these metrics, falling as a result,” he said.

“Gold’s declines have perhaps also been exacerbated by the strength of the gold price going into 2026 and possibly a desire amongst investors to liquidate profitable positions.”

Gold entered the year from an exceptionally elevated base, having surged to a record high above $5,595 per ounce in late January after a powerful multi-year rally driven by central bank buying, geopolitical hedging, and reserve diversification away from the dollar.

This means part of the current sell-off is less about a collapse in the long-term investment thesis and more about profit-taking in an overcrowded trade.

Iain Barnes, chief investment officer at Netwealth, said recent price moves have been unusually violent.

“International central banks seeking to diversify their reserves away from U.S. dollars may have started gold’s bull market in the past few years, but in the end the market ran out of new financial buyers and instead saw widespread profit-taking as wider uncertainty hit markets and the dollar rebounded,” he said.

Barnes also drew parallels with the 2008 financial crisis.

“In the first half of 2008, investors doubled down on the emerging market growth story, fueling commodity price increases alongside dollar weakness even as western economies hit the buffers,” he said.

“As the global financial crisis spread wider, global risk appetite collapsed and gold was hit alongside more productive commodities such as oil and copper as the dollar surged.”

He added: “This year, the market has again found where investors are most exposed: excessive positioning in gold as it was seen as the last remaining safe haven asset.”

That comparison helps explain why gold is falling even during wartime. In periods of market stress, investors often liquidate profitable assets to raise cash or meet margin calls elsewhere, temporarily overwhelming traditional haven demand.

Goldman Sachs, however, remains constructive on the medium-term outlook.

In a note on Monday, the bank said: “[But] we continue to forecast gold prices reaching $5,400/toz by end-2026, as central bank diversification continues, currently low speculative positioning normalizes, and the Fed delivers the 50bp of cuts our economists expect.”

The bank added that while risks remain skewed to the downside in the near term, the broader structural drivers remain intact.

“Over the medium term, risks are skewed to the upside if the Iran episode — together with broader geopolitical developments (e.g., Greenland, Venezuela) — were to accelerate diversification into gold and to weigh on perceptions of Western fiscal sustainability,” it said.

In effect, the market is witnessing a clash between short-term macro pressure and long-term structural demand. This explains why higher yields, a stronger dollar, and forced liquidation are currently dominating price action.

But beneath the current sell-off, central bank accumulation and geopolitical reserve diversification continue to provide support for the longer-term bull case. That tension is what makes this one of the most dramatic reversals in gold since 2008.

CNBC’s Jim Cramer on Monday urged investors not to abandon fundamentally strong stocks in the face of a market slide he says is being driven more by fear than by any real deterioration in corporate performance, warning that panic selling often creates some of the best buying opportunities.

Speaking on Mad Money, Cramer sought to push back against what he described as indiscriminate selling across high-quality names, particularly in the technology and cybersecurity sectors.

“Stocks go down for all sorts of reasons, some good, some bad. Lately we’ve had a lot of bad, and tonight I want to straighten some things out,” he said.

“Because a bad tape causes individuals to dump great stocks, usually when they should be buying more.”

His remarks came after a weak session on Wall Street that saw the major indices surrender early gains despite an initial rebound. The S&P 500 closed 0.39 per cent lower, the Nasdaq Composite dropped 0.73 per cent, while the Dow Jones Industrial Average eked out a 0.11 per cent gain.

The market weakness is unfolding against a backdrop of heightened geopolitical tension, rising oil prices, and persistent uncertainty over how long inflationary pressures from energy markets could weigh on risk assets. Cramer has recently warned that geopolitical volatility, particularly around the Middle East, is making market moves increasingly fragile.

But his central argument on Monday was that investors are misreading the impact of artificial intelligence on certain sectors.

Cramer pointed specifically to cybersecurity, where stocks such as Palo Alto Networks and CrowdStrike have come under pressure amid concerns that AI systems developed by private companies such as Anthropic could eventually displace traditional cyber defense platforms.

He flatly rejected that thesis.

“That is just dead wrong,” he said.

“In reality, the rise of AI should be a tailwind … for Palo Alto and CrowdStrike, because these same AI agents can be programmed by hackers to take over your network very easily. They are the vulnerability. Without the help of traditional cybersecurity, you’re more vulnerable than ever.”

That view goes to the heart of a broader debate on Wall Street over whether AI is a disruptor or an accelerator for legacy enterprise software firms.

In Cramer’s telling, AI does not eliminate the need for cybersecurity. Instead, it raises the stakes. As companies deploy autonomous agents and AI-driven workflows, the attack surface widens, creating fresh demand for endpoint protection, cloud security, and threat intelligence.

Enterprise security analysts have increasingly warned that generative AI tools can be weaponized by malicious actors to automate phishing, malware development, and network intrusion attempts at scale. In that context, established cybersecurity vendors may stand to benefit from higher spending rather than obsolescence.

Cramer also cited insider confidence as evidence that the market’s reaction may be misplaced. He pointed to a recent stock purchase by Nikesh Arora, who bought $10 million worth of shares.

“I don’t think a CEO would buy 10 million dollars’ worth of stock if he thought AI was an existential threat to the business model,” Cramer said.

The remark is significant because insider buying, particularly by chief executives, is often interpreted by investors as a strong signal of management confidence in future earnings and valuation.

Cramer then turned to Meta Platforms, whose stock has come under renewed pressure following recent legal setbacks. He argued that the market’s response has been excessive.

“I thought that the sell-off based on these lawsuits was strange,” he said, adding that such rulings are frequently challenged and often overturned on appeal.

Here, too, his argument is that investors are focusing too heavily on headline risk without adequately pricing in the long-term earnings power of the company’s advertising and AI businesses.

Meta remains one of the market’s most closely watched technology stocks, particularly as it continues to invest aggressively in artificial intelligence infrastructure and large-scale compute capacity.

Cramer’s broader message was that in markets dominated by fear, the disconnect between price action and business fundamentals can widen sharply.

“Sometimes stocks sell off for bad reasons, or fully bogus reasons, and at those moments, I’d rather be a buyer than a seller of CrowdStrike or Meta,” he said.

That stance is consistent with his recent commentary, where he has repeatedly urged investors not to confuse short-term market sentiment with long-term value, particularly in quality technology names.

The situation has presented an immediate challenge of ‘separating genuine fundamental risk’ from ‘fear-driven selling to investors.’ Cramer’s view is that the current market tape is doing more of the latter, and that for long-term holders, this may be a moment to accumulate rather than retreat.