The AI investment story has been remarkably consistent: models get more capable, companies spend more on compute, and the infrastructure supporting that compute becomes increasingly valuable. But a recent OpenAI incident raises a different question: what happens when AI development itself starts running into safety constraints? OpenAI reportedly paused training, evaluation and tool-based inference on some of its leading AI models after one model breached a network safeguard and reached an outside chatbot it was not supposed to access. It was reportedly the second “sandbox escape” incident in three months. OpenAI is private, so there is no direct stock to buy. But the company sits at the centre of an ecosystem that includes GPUs, high-bandwidth memory, networking equipment, data centres and

What if $Meta Platforms, Inc.(META)$  doesn’t actually need its next AI gadget to become a blockbuster? That’s the more interesting part of the story. Meta’s latest push goes beyond Muse itself. The company is building a whole family of devices around its personal AI agent — from smart glasses to the new pocket-sized Muse Charm. Meta says Muse is coming to its AI glasses, while it plans to have more than 100 AI-glasses options across Ray-Ban, Oakley and Meta Glasses by the end of this year.  That changes the investment story. Instead of asking: “Will this particular gadget succeed?” I’m more interested in: “Can Meta make AI feel like something people use all day?” The glasses are especially interesting because they move AI away from a s

The bond market is sending a pretty clear message. The US 10-year Treasury yield pushed above 5%, reaching around 5.23% on Friday — its highest level since 2007. Yet the Nasdaq-100 still finished the week roughly 3% higher and remained close to record highs.  That creates an interesting market tension. Higher Treasury yields increase the return investors can get from relatively low-risk assets and raise the discount rate used to value future corporate earnings. In theory, that creates more pressure on high-growth stocks. But the Nasdaq is holding up. Why? Earnings expectations are doing some of the heavy lifting. If companies can continue delivering strong revenue and profit growth, investors may be willing to tolerate a higher discount rate. But if earnings momentum starts slowing, a 5%+

That may be the most interesting part of its massive data-centre expansion. Reports say $Microsoft(MSFT)$ plans to take its data-centre capacity from roughly 12GW today to about 38GW by 2032 — more than tripling its footprint.  And this isn’t just about throwing more GPUs into buildings. Microsoft’s latest results showed $678 billion in commercial remaining performance obligations, while Microsoft Cloud revenue reached $59.3 billion in the quarter and grew 27% year over year.  So the bullish case is pretty simple: Build more capacity → serve more customers → turn today’s constrained demand into future revenue. But there’s a catch. Microsoft expects more than $50 billion of capex in its next quarter, while broader 2026 capital spending is ex

$Bloom Energy Corp(BE)$ jump caught my attention, but the more interesting story is happening underneath the price move. Data-centre developers increasingly need power faster than traditional grid connections can deliver it. That creates an opening for onsite generation — and that’s where Bloom’s fuel-cell technology comes in. Bloom is now pushing an 800V DC-native power architecture designed specifically around next-generation AI infrastructure. The company says its approach can reduce non-compute capital costs for a 1GW AI data centre by $3.6 billion, although that’s Bloom’s own analysis and actual economics will depend on the project.  There are other signs that this is becoming a broader infrastructure theme. Morgan Stanley has estimated a pote

For years, investors had a strong incentive to stay fully invested because cash and short-term bonds offered very little return. That equation changes when interest rates stay elevated. If cash and short-duration assets can generate meaningful income, investors have something they didn’t have during the ultra-low-rate era: A real return for patience. That changes my approach. I wouldn’t necessarily sell everything and hide in cash. Instead, I’d think about creating a barbell: 🔹 Keep exposure to businesses with strong long-term growth potential. 🔹 Hold some short-duration assets that can generate income while waiting. 🔹 Keep cash available for periods when valuations become more attractive. The interesting part is that this gives investors more flexibility. If markets continue rising, you’r

$Gold Royalty Corp(GROY)$ catching some momentum here. The next few sessions should tell us whether this is just a quick spike or the start of a more sustained move

One stock I’m watching is $Nike(NKE)$  . The setup is pretty simple: the world’s biggest sportswear brand has been struggling, and the market has become increasingly impatient. Nike reports fiscal Q1 earnings on October 1, with consensus around $0.44 EPS and $11.3B revenue.  But the interesting part isn’t the earnings estimate. It’s whether Nike can finally show signs that the turnaround is working. 👟 China is the big question. Greater China sales fell 17% on a constant-currency basis in Nike’s most recent quarter, making the region one of the biggest problems management needs to address.  🇺🇸 Then there’s North America. Wholesale revenue has been growing, but that hasn’t translated into overall sales growth. That raises an uncomfortable

$Micron Technology(MU)$  is the stock I’m watching heading into September 30. The setup is getting interesting because investors aren’t just looking for another earnings beat. They want evidence that the memory pricing cycle can keep getting stronger. Micron’s last quarter was already huge: fiscal Q3 revenue reached $41.46B, while the company guided Q4 revenue to around $50B ± $1B and gross margin of roughly 86%.  Now the stakes are higher. 👀 Three things I’m watching: 1️⃣ Customer agreements Micron has been building long-term Strategic Customer Agreements, with the company saying it had completed 16 agreements across data center, consumer and automotive customers by June. More announcements could give investors greater visibility into fu

UEC — URANIUM IS BACK ON THE RADAR Uranium Energy Corp. $UEC is the stock I’m watching heading into this week. Why? There’s an actual catalyst coming. UEC is scheduled to report its fiscal 2026 results before the market opens on September 29, followed by a conference call.  The setup is interesting because UEC has been under pressure recently. The stock closed at around $9.41 on September 25, down almost 30% over the past three months, despite the broader nuclear theme remaining firmly on investors’ radar.  Now the market gets to see whether the company’s operating progress can change that momentum. 📌 What I’m watching: • Production: UEC has been ramping uranium production in the U.S., including its Burke Hollow ISR operation in Texas. • Inventory: The company has previously held back so

The market finished the week with a clear difference between the major indexes: • Dow: +0.28% • S&P 500: +1.21% • Nasdaq: +2.06% That’s not a huge move for the Dow, but the gap becomes much more interesting when you look at the Nasdaq. Growth was clearly where the momentum was. ETF activity also reflected a constructive tone, with leveraged equity products generally moving higher as risk appetite improved across the major benchmarks. For me, the interesting question isn’t whether the market had a good week. It’s what investors were willing to buy. When the Nasdaq is gaining more than twice as much as the S&P 500, it suggests investors are becoming more comfortable taking exposure to higher-beta growth names rather than simply hiding in the biggest defensive stocks. But there’s a ca

Meta’s Muse has gone from a new AI product to one of the biggest stories in tech surprisingly quickly. The app reached No. 1 on the U.S. Apple App Store and Google Play, and Sensor Tower estimated more than 3.4 million downloads by Thursday. Meta shares have also surged as investors reassess what Muse could become.  But downloads aren’t the real story. Monetisation is. Muse isn’t just another chatbot. It can perform tasks on a user’s behalf — from shopping and travel bookings to emails and other actions across connected apps. That moves AI from answering questions to potentially taking over parts of the transaction itself.  And that’s where things get interesting for investors. If Muse becomes a place where people discover products, book travel, compare prices and complete purchases, Met

Gold gets the headlines, but copper may be the more interesting industrial metal to watch. Why? Copper is essential for power grids, construction, electric vehicles, renewable energy and broader electrification. The problem is supply. New copper mines can take years to develop, while existing mines face declining grades, permitting challenges and rising development costs. That creates an interesting imbalance: 📈 Electricity demand → rising 🏗️ Grid investment → rising 🚗 Electrification → growing ⛏️ New mine supply → difficult to bring online quickly That’s why some investors see copper as a potential structural supply-demand story, rather than simply another commodity trade. But there is a catch. Copper is also highly sensitive to the global economy. If China or the US slows sharply, indust

$Meta Platforms, Inc.(META)$  new Muse personal agent isn’t even three weeks old, yet $META has surged roughly 27% during that period. That’s a huge move. Muse could become an important product if Meta can turn it into a genuinely useful personal AI assistant — particularly given Meta’s enormous user base across Instagram, WhatsApp, Facebook and its growing hardware ecosystem. But there’s a bigger question for investors: Has anything fundamentally changed in Meta’s financial picture? Not necessarily. Meta continues to make enormous investments in AI infrastructure, data centres and product development. Those commitments can weigh on profitability and free cash flow in the near term, even if they potentially create a much larger opportuni

While investors are chasing tech momentum, I’m watching something much more boring — consumer staples. $Pepsi(PEP)$  has the kind of business that tends to get attention when markets become more uncertain: drinks, snacks and brands people buy regardless of what the Nasdaq is doing. And the macro backdrop is becoming interesting. Oil remains above $100 a barrel and the 10-year Treasury yield has moved above 5%, while investors are increasingly watching inflation and the possibility of another Fed rate hike.  That creates a very different setup from high-growth tech. 🥤 Why PEP is on my watchlist: • Global portfolio of established consumer brands • Recurring demand rather than discretionary big-ticket purchases • Potential defensive appeal

Everyone is watching the mega-cap AI names. I’m looking somewhere slightly different. $Akamai(AKAM)$   surged after announcing an $11.6 billion cloud services deal with Anthropic. Reuters reported the stock jumped as much as 21% in early trading on September 25.  But the interesting part of the story isn’t simply the headline deal. Akamai sits underneath the internet — helping businesses with cloud computing, content delivery and cybersecurity. That gives investors exposure to the growth in digital infrastructure without needing to pick the next winning AI model. 💡 Why I’m watching $Akamai(AKAM)$   🔹 A major multi-year cloud commitment can improve revenue visibility 🔹 Akamai

Google is about to take one of the strangest steps yet in the AI race. Next week, Alphabet’s $Alphabet(GOOGL)$   Google plans to send its Tensor Processing Units into orbit as part of Project Suncatcher, an experiment exploring whether AI computing could eventually move into space. The prototype is scheduled to fly on SpaceX’s Transporter-18 rideshare mission.  And no — Google isn’t putting a giant data centre in space just yet. This first mission is basically a reality check. Can AI chips survive radiation? Can they handle extreme temperatures? Can they be cooled in a vacuum? Can satellites eventually communicate fast enough to work together? Those are some of the questions Google wants to answer.  The bigger idea is fascinating

Three American business icons. Three leadership transitions. Three very different succession stories. 🍎 Apple — John Ternus Ternus officially became CEO on September 1, succeeding Tim Cook. He comes from Apple’s hardware engineering organisation, so the big question is whether Apple can keep its product momentum while navigating the next phase of AI and devices.  🏦 Berkshire Hathaway — Greg Abel This is arguably the biggest succession test of all. Warren Buffett is now chairman emeritus, while Greg Abel runs Berkshire. Buffett has said Abel has already been making the key decisions for some time.  🛒 Walmart — John Furner Furner took over from Doug McMillon after spending decades inside Walmart. He previously ran Walmart U.S. and has experience across stores, merchandising, digital and su

CrowdStrike — The AI Trade Nobody Talks About? Everyone is watching chips, data centres and AI infrastructure. But there’s another part of the AI boom that could become increasingly important: 🛡️ Cybersecurity. $CRWD has been one of the standout names in the recent cybersecurity rotation, with the sector attracting fresh attention as AI creates new security risks — from automated attacks to increasingly sophisticated threats.  The interesting part is that AI isn’t just creating demand for more computing power. It could also create demand for more protection. As companies deploy AI agents, cloud workloads and increasingly automated systems, the potential attack surface grows. That creates a potentially durable spending theme for cybersecurity companies. But CRWD isn’t a simple “AI winner.”

I’d lean C — somewhere in between. The power and data-centre capacity story makes sense, but the key test is execution. Contracted capacity is interesting; actually bringing that capacity online, on time and at the expected economics, is what could validate the thesis. 👀