AI daily – China’s moonshot fires another shot across the AI bow
The AI trade was already beginning to sag under the weight of its own ambition. Capital spending keeps climbing, hyperscaler balance sheets are being stretched, and investors are asking harder questions about whether the economics will ever catch up with the infrastructure bill. Then China’s Moonshot AI unveiled Kimi K3, and another DeepSeek-style tremor ran through US technology stocks.
The model does not need to dethrone OpenAI or Anthropic to matter. It only needs to prove that frontier-level capability is becoming cheaper, more widely available and increasingly difficult to monopolise. That is the real threat to the old AI script.
Kimi K3 reportedly sits near the top of the current model rankings, with strong results across reasoning, coding and long-context workloads.
It is also being positioned as a large open-weight model with native text, image and video capabilities, alongside aggressive API pricing that undercuts many premium closed systems.
That is where the market’s nerves begin to fray.
The US AI complex has spent the past several years trading on the belief that enormous data-centre investment would create an almost unassailable moat.
The logic was simple: whoever controlled the largest models, the most advanced chips and the deepest pools of computing power would control the economics of the next technological era.
But each capable Chinese model arriving at a lower price chips away at that assumption. If customers can achieve comparable results without paying premium rates to the leading US platforms, then the return on hundreds of billions of dollars in AI infrastructure becomes harder to defend. The concern is no longer whether demand for AI exists. It clearly does. The concern is who ultimately captures the margin.
Kimi K3 is also being pitched as more than another chatbot. Moonshot claims the model can handle complex engineering tasks, including chip design, optimisation, verification and simulation, while also building software tools that compete with parts of established developer ecosystems. Those claims will need proper testing, but the direction of travel is difficult to ignore.
AI capability is moving beyond polished demonstrations and into practical automation.
That matters for equities because the AI trade is becoming less forgiving. Semiconductors, memory stocks and hyperscalers have all benefited from the assumption that spending would rise in a straight line and that every new model would require more premium hardware. Lower-cost alternatives complicate that story. They raise the possibility that intelligence becomes more efficient even as infrastructure supply continues to surge.
In other words, the market may be discovering that more AI does not automatically mean more profit for every company carrying an AI label.
Moonshot is wasting little time capitalising on the attention. The three-year-old company is reportedly preparing for a Hong Kong listing within six months while completing a funding round at a valuation near $30 billion. Annualised revenue has also climbed sharply, supported by subscriptions and enterprise services.
The IPO race is now becoming part of the AI arms race itself. Chinese and US model developers are all moving toward public markets, where investors will finally get a clearer look at revenue quality, cash burn, infrastructure costs and the gap between technological excitement and commercial return.
For markets, Moonshot’s rise is less about whether Kimi K3 is definitively the world’s best model and more about what it says about scarcity. The AI boom was built on the idea that frontier intelligence would remain rare, expensive and concentrated in a handful of US companies.
China keeps suggesting otherwise.
And every time another low-cost model crosses the frontier, the premium attached to the old story becomes a little harder to justify.
Markets daily – Morgan Stanley says the cycle still has road, but this is a market for Gamma over Theta
Morgan Stanley’s Andrew Sheets is reaching into the history books for a map of the current market, but not because he believes the past is preparing to repeat itself line for line. Markets rarely offer that kind of courtesy. The variables shift, the actors change, and the same macro ingredients can produce a very different meal. Still, some periods rhyme closely enough to be useful, and Sheets continues to see 1997–98 and 2005–06 as the best guides to the road ahead.
The common thread is a cycle that still has distance to travel, rising corporate aggression, and a macro backdrop steady enough to keep risk appetite alive. Morgan Stanley’s broader conclusion is that equities should continue to outperform credit, while investors are better served owning volatility in rates and FX rather than harvesting carry and hoping the road stays smooth.
The first pillar of the argument is the sharp revival in corporate spending. For years, investors became accustomed to a capital-light economy where businesses preferred buybacks, software and balance-sheet efficiency over bricks, power, factories and machinery. That era is being challenged. Morgan Stanley notes that US capital expenditure growth among the Russell 1000 averaged roughly 7% annually from 2010 through 2025, before accelerating by 33% in 2025. The bank forecasts another 23% increase in 2026 and 26% in 2027.
AI remains the main engine, but it is not driving alone. Energy infrastructure, more favourable tax treatment and a broader investment cycle across Asia are adding fuel. The important point is that this is no longer a narrow technology story. Capital spending is becoming a global macro force.
M&A is also waking from a long sleep. Global deal volumes had fallen to historic lows relative to the size of the economy in early 2024, but announced transactions are now up 64% from a year ago. Funding markets remain open, regulation is becoming friendlier, and companies are making up for lost time. In Morgan Stanley’s historical analogues, corporate aggression continued to intensify before the credit cycle finally peaked. That suggests the clock is not yet at midnight.
The catch is that more aggression also means more supply. Companies investing, borrowing and buying assets can keep equity earnings moving higher, but the additional debt issuance may pressure credit spreads. That is why Morgan Stanley prefers equities over credit. The growth engine remains alive, but bond investors may be asked to absorb more paper without being paid enough for the privilege.
The macro backdrop also fits the comparison. Core inflation, unemployment, the US 10-year yield and the shape of the curve sit in a range that resembles those earlier periods. Growth remains respectable, rates are not yet crushing activity, and the expansion does not look exhausted. Add deregulation across banking, insurance and savings systems, and the ingredients for further risk-taking remain in place.
The larger question is how to own the cycle.
The late 1990s rewarded investors who stayed long risk but also owned protection. It was a ride-the-whirlwind market, where the upside remained powerful but the path was increasingly violent. The mid-2000s were kinder to volatility sellers, with theta slowly doing the heavy lifting as markets ground higher.
Sheets prefers the former template. Call it gamma over theta.
The message is not to abandon risk, but to stop assuming the journey will be calm. Corporate spending is accelerating, policy visibility is fading, and central banks have less room to guide markets with confidence. Morgan Stanley still sees road ahead for equities, but the better seat may belong to investors who remain long the cycle while keeping optionality close at hand.
Europe daily – China is selling more into Europe, but the equity damage is less obvious than it looks
Europe’s trade relationship with China is becoming steadily more one-sided. Chinese companies now account for roughly 23% of EU imports, up from 21% two years ago, while Europe’s share of exports into China has fallen sharply since 2020. Put simply, China is selling more into Europe while buying less from it, and that is becoming a growing competitive headache for the region’s industrial base.
The pressure is no longer confined to cheap consumer goods or low-end manufacturing. Chinese companies are gaining ground across automobiles, medical technology, chemicals and other industries that once sat comfortably inside Europe’s competitive moat. Better products, aggressive pricing and enormous manufacturing scale are allowing Chinese firms to push deeper into markets that European companies once treated as home turf.
For manufacturers, the challenge is clear. They are being squeezed at both ends of the trade route. Access to the Chinese market is becoming less rewarding, while competition inside Europe is becoming more intense. That is a difficult equation for companies already dealing with high energy costs, heavier regulation and slower domestic growth.
Yet the stock-market impact is less straightforward than the trade numbers suggest.
Investors often treat rising Chinese exports as a direct negative for European equities, but the major indexes are not simple reflections of the manufacturing economy. Their composition has changed. Technology hardware, semiconductor equipment, healthcare, financial services, telecommunications and tourism now carry far more weight than they once did, and many of these sectors are either benefiting from structural growth or remain relatively insulated from direct Chinese competition.
That distinction matters. Europe’s automakers may be losing ground, particularly in electric vehicles, but the weakness of one industry does not automatically sink the broader market. Semiconductor equipment and other high-value technology businesses have expanded rapidly, helping offset some of the pressure in traditional industrial sectors. Europe may be losing share in parts of the old economy while still producing companies that sit near the tollbooths of the new one.
The regional market also contains large sectors where China is not the primary threat. Banks are driven more by domestic credit conditions, interest rates and capital returns. Telecom companies depend on local networks and regulation. Tourism benefits from travel demand rather than export competitiveness. These businesses can continue generating earnings even as Chinese manufacturers capture a larger slice of the goods market.
There is also an important difference in how shareholder returns are treated. European companies have become more disciplined on dividends, buybacks and capital allocation, while many Chinese businesses remain focused on expansion, capacity and market share. That can create a strange outcome where China wins the industrial contest but European equities still deliver the better investment return.
The numbers reflect that tension. The STOXX 600 has gained about 8% this year through mid-July, while China’s CSI 300 has risen only around 1.5% in local-currency terms. Europe is clearly facing a competitive challenge, but the market is not waiting for every factory to wave a white flag.
The real risk is concentration beneath the index. The headline benchmark may remain resilient while the damage gathers in autos, chemicals and other manufacturing-heavy corners of the market. Europe is not a single trade, and investors need to separate the companies exposed to China’s export machine from those sheltered by services, technology leadership, or domestic demand.
China is taking a larger bite out of Europe’s industrial lunch. But when it comes to European stocks, the menu is broader than many investors assume.
Dark Side of the Boom in the news
Higher crude prices have revived concerns that inflation could remain elevated and complicate the path to lower interest rates, but some analysts argue the broader economic backdrop is becoming more supportive.
“Markets are once again being forced to trade two seemingly contradictory stories on the same screen,” said Stephen Innes of SPI Asset Management.
While the renewed rise in oil prices has injected a fresh geopolitical risk premium into markets, he said cooling underlying US inflation and a softer labour market suggested the energy shock would not necessarily trigger a new cycle of broad-based inflation.
Instead, the biggest risk would come if elevated oil prices persist long enough to erode household spending and weigh on economic growth.
With more than 25 years of experience, Stephen has a deep-seated knowledge of G10 and Asian currency markets as well as precious metal and oil markets.
