Technology investors are facing an uncomfortable combination this week.
$Invesco QQQ(QQQ)$ fell 1.20% on Friday, significantly more than $SPDR S&P 500 ETF Trust(SPY)$ ,which declined 0.36%. This suggests investors are becoming more cautious about growth stocks, particularly those whose valuations depend heavily on future earnings.
But the bigger issue is not simply another technology sell-off.
Two important events could influence market sentiment this week: Federal Reserve Chair Kevin Warsh’s discussion about AI productivity and employment, followed by Nvidia’s earnings announcement on Tuesday.
Together, they raise a question that could shape the technology market for months:
Will AI eventually generate enough productivity and profits to justify today’s enormous investment, or will higher borrowing costs and inflation slow the entire AI boom?
I think investors need to look at both sides of this argument.
1. Why is the Fed paying so much attention to AI?
On Monday, November 16, the Federal Reserve and San Francisco Fed are holding a Productivity and Jobs Task Force conference.
Kevin Warsh is scheduled to deliver the opening remarks, with discussions covering AI adoption, productivity, employment and the wider economic consequences of advanced AI technology.
This is important because AI could influence monetary policy in two very different ways.
On the positive side, AI could help businesses produce more with fewer resources. Companies could automate repetitive tasks, improve efficiency, reduce operating costs and increase output without hiring proportionally more employees.
If productivity improves significantly, businesses may be able to grow earnings without creating the same inflationary pressures that normally accompany rapid economic expansion.
That would be positive for technology stocks.
However, there is another side.
Building AI infrastructure requires enormous investment in data centres, GPUs, electricity, networking equipment and construction.
These investments are happening now, while the productivity benefits may take years to appear.
In the meantime, demand for electricity, skilled labour, computing equipment and infrastructure could push costs higher.
This creates a difficult situation for the Fed: AI may eventually help reduce inflation, but its development could contribute to inflation in the short term.
2. Inflation is becoming a bigger problem for growth stocks
October’s US CPI figures showed headline inflation increasing 0.4% month-on-month, while core CPI increased 0.3%.
Although these numbers may look relatively small, persistent monthly increases can become a problem when inflation is already above the Fed’s 2% target.
The concern is that inflation may remain stubborn even if economic growth begins slowing.
If the Fed needs to keep interest rates higher for longer, technology companies could face several challenges.
First, higher borrowing costs make expensive data-centre projects more costly.
Second, investors may demand higher returns before paying premium valuations for growth stocks.
Third, future earnings become less valuable when discounted at higher interest rates.
This is particularly important for QQQ because it has significant exposure to large technology and growth companies.
A company can report excellent earnings and still see its share price decline if investors believe its valuation has become too expensive.
That is why I would pay close attention to Treasury yields following Warsh’s remarks.
3. What should investors listen for from Kevin Warsh?
I would focus on three areas rather than expecting one speech to determine the Fed’s next interest-rate decision.
• Productivity: Does Warsh believe AI can deliver meaningful productivity improvements in the near future, or will the benefits take longer to materialise?
• Inflation: Does he acknowledge that data-centre construction, energy demand and AI infrastructure spending could create additional inflationary pressure?
• Employment: How does the Fed assess the possibility that AI could increase productivity while reducing demand for certain types of labour?
The distinction between short-term costs and long-term benefits matters.
If Warsh sounds optimistic about AI productivity but cautious about near-term inflation, investors may interpret that as a mixed signal.
If he emphasises inflation risks and the need to maintain restrictive monetary policy, technology stocks could face further pressure.
Importantly, this conference is intended to gather evidence and inform the Fed’s thinking. It is not an announcement of a new interest-rate decision.
4. Nvidia earnings could determine the direction of the AI trade
The second major event is Nvidia’s earnings announcement on Tuesday, November 17, after the US market closes.
Nvidia previously reported quarterly revenue of $96.2 billion, with Data Centre revenue contributing approximately $89 billion.
For the next quarter, management guided for approximately $108 billion in revenue.
That is an enormous expectation.
However, this earnings announcement comes at a particularly sensitive time.
Investors are already dealing with leadership uncertainty following Jensen Huang’s departure, alongside questions about AI infrastructure spending and competition from AMD and custom AI chips.
I would focus on four things:
1. Revenue guidance: Can Nvidia meet or exceed its approximately $108 billion forecast?
2. Data Centre demand: Are major cloud providers continuing to commit billions to AI infrastructure?
3. Gross margins: Can Nvidia maintain profitability as production expands?
4. Forward outlook: Is demand sustainable, or are customers becoming more cautious about their capital expenditure?
I would also pay attention to management’s comments about customer commitments and delivery schedules.
A strong earnings beat may provide short-term relief, but investors need confidence that demand will continue beyond the current quarter.
If Nvidia delivers strong results and positive guidance, it could help restore confidence across the semiconductor sector.
However, if the results are good but the outlook disappoints, the wider AI trade could remain under pressure.
5. What does this mean for QQQ? Three possible scenarios
I see three potential outcomes.
Bullish scenario: AI productivity and earnings win
Warsh acknowledges the long-term productivity benefits of AI without signalling an immediate need for more aggressive monetary tightening.
Treasury yields stabilise or decline, while Nvidia delivers strong earnings and confident guidance.
This combination could encourage investors to return to growth stocks.
QQQ could recover as technology investors regain confidence in the AI investment cycle.
Neutral scenario: Good earnings, but rates remain a concern
Nvidia beats expectations, but Warsh remains cautious about inflation and the timing of AI productivity gains.
The market may initially rally on Nvidia’s results, but higher Treasury yields could limit further upside.
In this situation, I would expect greater differences between individual technology stocks rather than a broad recovery across the sector.
Bearish scenario: Inflation and AI spending concerns dominate
Warsh emphasises the inflationary risks associated with AI investment, while Nvidia provides weaker-than-expected guidance.
Treasury yields move higher and investors reduce exposure to expensive growth stocks.
QQQ could experience another round of selling, particularly among companies whose valuations depend on earnings expected several years into the future.
A 25–50 basis-point December rate increase is one possible tightening scenario being discussed in the market, not a confirmed Fed decision. The actual outcome will depend on incoming economic data and policymakers’ assessments.
6. How would I position my portfolio this week?
Personally, I would avoid making an aggressive move in either direction before these events.
I would use a more defensive approach while keeping exposure to the long-term AI opportunity.
My illustrative allocation would be:
50% – Core diversified ETFs
I would keep the largest allocation in diversified investments rather than relying entirely on technology stocks.
This provides exposure to companies across different sectors and reduces the impact of a disappointing earnings announcement from one major technology company.
20% – Established technology companies
I would maintain exposure to established businesses with strong cash flows, competitive advantages and proven demand.
However, I would be selective about valuations rather than assuming every large technology company is automatically a safe investment.
10% – Higher-growth AI opportunities
I would reserve a smaller portion for companies such as Nvidia, AMD and Broadcom.
These businesses have significant long-term opportunities, but their share prices can be particularly sensitive to earnings expectations and market sentiment.
20% – Cash reserves
I would keep some cash available rather than investing everything immediately.
If QQQ falls further following Nvidia’s earnings, having cash available provides flexibility to add to positions gradually.
These percentages are an illustrative strategy, not a recommendation for every investor.
My conclusion: I am not ready to abandon the AI trade, but I am becoming more selective.
I believe AI has the potential to transform business productivity and create substantial long-term economic value.
However, investors need to remember that a promising technology does not automatically make every company in that sector a good investment at any price.
The market is now asking two important questions.
Can AI companies turn enormous capital expenditure into sustainable profits?
And can the Fed allow the economy to benefit from AI productivity without allowing inflation to accelerate in the meantime?
For QQQ, I think the combination of Treasury yields, Nvidia’s forward guidance and the Fed’s assessment of AI productivity will matter more than any single headline.
I would rather maintain some cash and wait for confirmation than rush to buy every dip.
The AI revolution may still have years to run, but that does not mean technology stocks cannot experience a painful correction along the way.
For discussion purposes only.
Comments