Why People Missed Dot-Com, Data Centers Next Door, Crypto’s Threat to Power, Why Flights Stay Expensive, Deflating the Portfolio Balloon, AI Boom or Bust, Simple vs. Compound Loans & More

Did you ever wonder why so many people didn't get out before the dot-com crash?

It's an important question to ask yourself, especially if you believe you'll know exactly when to get out before any potential correction in today's AI and semiconductor stocks.

The reality is that the dot-com bubble burst only 25 years ago. Human nature hasn't changed since then. Investors today are no smarter than investors were back then, and the same emotions that drove the bubble are showing up again. There were four major reasons so many people lost money during the tech bust.

The first was that investors stopped focusing on earnings and price-to-earnings ratios. Instead, they justified sky-high valuations by looking at metrics like website traffic, page views, click-through rates, and the number of "eyeballs" on a screen. The assumption was that if revenue kept growing, profits would eventually follow. Many ignored the reality that businesses also have expenses, competition, and execution risk.

The second reason was FOMO or the fear of missing out. Between 1995 and 2000, the Nasdaq surged roughly 400%. As people watched friends, coworkers, and investors make fortunes on tech stocks and IPOs, more and more money poured into the market. Institutional investors and retail investors alike stopped worrying about valuations. They simply saw stocks going up and didn't want to miss the ride.

The third reason was the belief that "this time is different." You heard it everywhere: "You just don't get it. This is the new economy." Investors argued that traditional valuation metrics no longer mattered because the only thing that counted was gaining market share. Profitability could always come later.

The fourth reason was the assumption that capital would never dry up. Few investors paid attention to where companies were getting their money. Many businesses were surviving on venture capital rather than sustainable profits. When funding slowed and investors became more selective, those companies had no profitable business model to fall back on. Many quickly went bankrupt.

At the peak of the bubble, investors stopped asking basic questions. What am I paying for this company's earnings? What am I paying for its cash flow? In many cases, there weren't any. Yet investors convinced themselves the speculative frenzy would continue indefinitely.

The biggest lesson is a humbling one. We like to believe we'll recognize the top and get out before everyone else. But investors in 2000 believed the same thing. Human psychology hasn't changed, which is why bubbles continue to repeat throughout history.

Don’t Build That Data Center in My Backyard

The race to build AI infrastructure is running into an obstacle that many investors probably didn't see coming: local communities.

Across the country, residents are protesting and filing lawsuits to stop new AI data centers from being built in their neighborhoods. One of the biggest concerns is something most people never think about, the constant noise. Data centers operate around the clock, with cooling fans, chillers, and backup generators creating a continuous hum 24 hours a day. That may not sound like a major issue until you have to live next to it.

New York has become one of the focal points of this debate. While the state has plenty of available land for development, many communities are pushing back. Governor Kathy Hochul is even considering legislation that would place a moratorium on the construction of large data centers in certain areas.

Public opinion reflects that growing resistance. According to recent polling, 44% of Americans oppose additional data center construction, while only 21% support it. When the question becomes more personal and whether people would support a data center being built in their own community, opposition jumps to 57%, while support falls to just 14%.

Residents also question the long-term economic benefits. Building a data center may create thousands of construction jobs, but once the facility is complete, permanent employment may fall to just 100 to 200 workers. At the same time, these facilities consume enormous amounts of electricity. In some regions served by smaller utilities, a single data center could account for as much as 25% of total power demand, raising concerns about higher electricity costs and increased strain on the grid.

The political landscape is becoming more challenging. Lawmakers in states including Arizona, Illinois, and Ohio have restricted or eliminated tax incentives that were previously used to attract data center investment.

Even the companies building this infrastructure recognize the growing risk. The hyperscalers are expected to spend nearly $1 trillion on AI infrastructure this year, but increasing public opposition could slow those plans. Nebius Group, for example, warned in its 2025 annual report that rising resistance to data center projects in certain communities could become a headwind for future expansion.

Investors have spent a great deal of time focusing on AI demand, chips, and software. However, another risk is emerging that deserves attention: if communities continue saying, "Not in my backyard," the pace of AI infrastructure growth may not be as smooth as many expect.

Is Crypto Weakening One of America's Most Powerful Weapons?

One of the United States' greatest geopolitical advantages isn't its military, it's the U.S. dollar.

Roughly 90% of global foreign exchange transactions involve the U.S. dollar. That dominance gives the United States enormous leverage. When the U.S. imposes financial sanctions and cuts countries off from the dollar-based financial system, it becomes far more difficult for them to conduct international trade, finance military operations, or access global markets.

That advantage is beginning to erode. Countries that have long opposed the United States such as Russia, Iran, and North Korea are increasingly turning to cryptocurrencies to bypass traditional financial channels. According to reports, their use of virtual currencies for cross-border transactions surged from roughly $12.5 billion in 2024 to more than $100 billion in 2025.

Crypto gives sanctioned nations another way to move money. It can be used to purchase drones, weapons, military components, and fuel, while also helping finance operations such as smuggling oil and paying suppliers outside the traditional banking system.

North Korea has become one of the world's most aggressive crypto thieves, using hacking and other cybercrimes to steal digital assets that can then be converted into funding for its military and weapons programs.

Part of the challenge is that cryptocurrency wallets are identified by long strings of letters and numbers rather than names. While blockchain transactions are publicly visible, identifying the person or organization controlling a wallet can be extremely difficult without additional intelligence. That makes enforcement of financial sanctions much harder.

Even terrorist organizations such as Hamas have, at times, solicited donations in cryptocurrency, illustrating how digital assets can be used to circumvent traditional financial controls.

This is why I believe cryptocurrency has become more than just an investment story, it has become a national security issue.

If Bitcoin and other cryptocurrencies were to experience a significant decline in value, it would reduce the purchasing power of those holding large crypto reserves, including sanctioned actors that rely on digital assets. While it would not eliminate their ability to use crypto, it could make this alternative financial system less effective and increase the relative importance of the dollar-based financial system.

The stronger the role of the U.S. dollar in global commerce, the more effective financial sanctions remain as a non-military tool of foreign policy. With cryptocurrencies becoming more widely adopted, policymakers will need to consider the risk of weakening one of America's most effective forms of economic leverage.

Even with oil off its recent peak, you still may not see cheaper airline tickets

You might assume that with the decline in oil prices, jet fuel costs are also declining, and airlines will pass those savings on to travelers through lower ticket prices. Oil and jet fuel prices have indeed come down, but don't expect airlines to slash fares anytime soon.

The reason is simple: demand remains strong. Even after airlines raised fares eight times since the start of the conflict in the Middle East, analysts say the average round-trip domestic ticket climbed roughly 19% to about $638 yet demand barely changed. In other words, consumers have shown they are willing to pay higher prices to travel. If people keep buying tickets, airlines have little incentive to lower fares and give up those higher profit margins.

Supply is also likely to remain constrained. Airlines aren't rushing to add flights because keeping capacity tight helps support higher ticket prices. The bankruptcy and downsizing of low-cost carriers such as Spirit Airlines has also reduced competition on many routes, making it easier for the remaining airlines to maintain pricing power.

To be fair, airline pricing should be viewed over a longer time horizon. From 2019 through 2025, overall consumer prices rose about 26%, while average airfares actually declined roughly 3.5%. So, despite the recent increases, airline tickets are still relatively inexpensive compared with the broader rise in inflation over the past six years.

The bottom line is that lower fuel costs alone don't guarantee lower ticket prices. As long as travel demand remains healthy and airlines keep capacity in check, consumers may not see much relief at the checkout screen.

Letting Air Out of the Investment Portfolio Balloon Before It Pops

At one point or another, we've all seen a balloon inflated until it finally bursts. The same thing can happen to an investment portfolio.

Watching your portfolio grow is exciting, but every investor knows that markets don't go up forever. The challenge is that no one knows exactly when a portfolio has become too inflated. One of the biggest reasons investors refuse to sell is simple: they hate paying taxes. Believe me, I dislike paying taxes just as much as anyone else. But you should never let the tax bill dictate your investment decisions.

Sometimes the smartest move is to relieve some of the pressure in your portfolio before the market does it for you. There are two simple ways to accomplish this: trim oversized positions and sell investments that have become significantly overvalued.

The first strategy is reducing concentration risk. If you review your portfolio and discover that a single stock has grown to 10% or 12% of your total assets, it may be time to trim that position back to 7% or 8%. Yes, you'll likely owe capital gains taxes, but you'll also be reducing the risk that one investment can have an outsized impact on your portfolio if it suddenly declines.

The second strategy is selling investments that have exceeded your target price and can no longer be justified based on their fundamentals. If the valuation has become stretched and the company's earnings outlook no longer supports the stock price, it may be time to take profits. Again, you'll probably owe taxes on the gain, but remember that capital gains are generally taxed at favorable rates. More importantly, paying a 20% or 25% tax on your profit is often far less painful than watching the entire investment lose 20% or more in value. That 20% decline occurs on the entire position rather than just the gain.

No strategy is perfect. You may trim a position only to watch it continue climbing for another year or two. That's part of investing. Risk management isn't about perfectly timing the top, it's about ensuring that no single investment or sector can seriously damage your long-term financial plan.

Consistently following a disciplined, conservative approach won't always maximize returns during bull markets, but it can significantly reduce risk over a full market cycle. When the next major correction inevitably arrives, your portfolio should be positioned to withstand it. That makes it far easier to stay invested, avoid emotional decisions, and continue building wealth instead of panic-selling after the damage has already been done.

Successful investing isn't just about finding great investments. It's also about knowing when to reduce risk. Sometimes, letting a little air out of the balloon today is the best way to keep it from popping tomorrow.

Is AI creating the next memory boom... or setting up the next bust?

SK Hynix just pulled off the largest foreign ADR listing in U.S. history, pricing its American depositary receipts at $149 and raising $26.5 billion. That isn't just a fundraising event, it is fuel for one of the most aggressive semiconductor expansion plans the industry has ever seen.

The company is pouring money into new factories, equipment, and advanced packaging capacity around the world. In the United States, SK Hynix is building its first manufacturing facility, a $4 billion advanced packaging plant in West Lafayette, Indiana, expected to be completed in 2028.

Back home in South Korea, the spending is even more staggering. SK Hynix plans to invest up to $720 billion expanding memory production, including a $390 billion semiconductor cluster in Yongin. The company has also committed roughly $7.8 billion by the end of 2027 for additional extreme ultraviolet (EUV) lithography machines, the highly specialized tools needed to manufacture cutting-edge HBM chips. These machines cost as much as $400 million each, are in extremely limited supply, and are only produced by ASML.

The company is even accelerating its expansion timeline by more than a decade, with four new fabrication plants now expected to be completed by 2033.

The question investors should be asking isn't whether AI demand is real. It clearly is. The real question is whether the industry is repeating a familiar pattern. Memory has always been one of the most cyclical businesses in technology. Every major technology revolution from the dot-com boom, to smartphones, to cloud computing created a surge in demand for memory chips. Manufacturers responded by rapidly expanding production. Eventually supply caught up, prices collapsed, profits disappeared, and investors who arrived late learned just how brutal the memory cycle can be.

Today feels different... but that is often what every cycle feels like while it is happening.

SK Hynix's market value has increased more than sevenfold over the past year as AI infrastructure spending has created a shortage of HBM. Revenue nearly tripled between 2023 and 2025 to roughly $65 billion, and Wall Street expects sales to surge again to approximately $235 billion in 2026.

Those are incredible numbers. But when major memory producers start announcing massive capacity expansions, history suggests investors should at least consider what happens when today's shortage eventually becomes tomorrow's surplus. AI may create years of strong demand for memory, but the semiconductor industry has a long history of building too much capacity just as demand begins to normalize. The opportunity is enormous, but so is the risk if history repeats itself.

Financial Planning: Simple vs Compounding Interest Loans

Many people assume that choosing a simple interest loan over a compound interest loan will dramatically reduce the amount of interest they pay, but in most real-world lending situations, the difference is minimal. The reason is that the power of compounding only becomes significant when a balance grows over time because interest is being added to the principal. With most consumer loans, borrowers either make interest-only payments that keep the principal balance unchanged or make payments that reduce the principal over time. In either case, the interest charged during each payment period is based on the outstanding loan balance at that time, not on an ever-growing balance. Since the loan balance is remaining the same or steadily declining rather than increasing, there is little opportunity for “interest on interest” to accumulate. While compounding can become important if unpaid interest is capitalized and added to the loan balance, that is the exception rather than the rule. For most mortgages, HELOCs, auto loans, personal loans, and similar debt, borrowers should focus far more on the interest rate than on whether the loan is described as using simple or compound interest.

Too Many People Are Using Target Date Funds in Their 401(k)

For years, we've discussed the drawbacks of target date funds, including their higher fees and one-size-fits-all approach. Despite those concerns, they remain incredibly popular because they are simple and require very little effort from the investor. According to Vanguard, 61% of 401(k) participants invest in target date funds.

On the surface, they sound like the perfect solution. If you plan to retire around 2045, you simply choose the 2045 Target Date Fund and let it manage your investments. The fund automatically adjusts your portfolio over time, gradually reducing your exposure to stocks and increasing your allocation to bonds as you approach retirement.

Many investors don't realize how significant that shift can be. By the target retirement date, a target date fund may hold around 50% of its assets in bonds. The adjustments don't stop there. Reaching the target year doesn't mean the fund is liquidated or that you receive your money. Instead, the fund continues along its glide path and could increase its bond allocation to 70% or even 80% over the following years.

That approach may have made sense decades ago, but retirement looks very different today. Many people will spend 20 years or more in retirement. Over that length of time, maintaining enough exposure to stocks can be critical to helping your portfolio grow and keep pace with inflation. A portfolio that becomes too conservative too quickly may struggle to provide the long-term growth many retirees need.

Another limitation is that target date funds only manage the assets inside your 401(k). They don't take into account your IRAs, brokerage accounts, pensions, real estate, or other investments. As a result, your overall portfolio allocation could end up being far different than what is appropriate for your financial goals.

The convenience of target date funds is appealing, but convenience shouldn't replace planning. A successful retirement requires understanding how your money is invested, estimating what your portfolio could be worth when you retire, and developing a strategy for how those assets will be invested throughout retirement, not just until you reach it.

Is That Really Your Son or Daughter Calling You?

You know your children's voices. You talk to them regularly. Then one day you get a frantic phone call from your son or daughter. They tell you they've just been in a serious accident. They need $15,000 immediately or they're going to jail. They tell you exactly how to send the money. Without hesitation, you wire the funds because you want to help your child.

Unfortunately, you have just been scammed by AI. AI-powered scams are exploding. Reports show AI-related fraud surged more than 1,200% in 2025, and at the current pace, losses from AI scams in the United States could reach $40 billion annually by 2027. Another study found that one in four adults has already experienced an AI voice scam.

Your first reaction may be, "That could never happen to me. I don't post anything on social media." But the problem may not be your online presence. It's your children.

Many people regularly post videos on social media, and today's AI only needs about three seconds of someone's voice to create a convincing clone. Once scammers have that sample, they can make it sound like your son or daughter is saying almost anything.

So how do you protect yourself? If you receive an emergency call asking for money, don't panic. Before sending anything, ask a question that only you and your child would know the answer to. Make it something that has never been shared publicly.

For example, ask about a funny childhood memory that only the two of you remember. Don't use information like birthdays, graduation dates, wedding dates, or other facts that could be found online or in public records. Remember with all these data centers there is so much information that is being obtained and saved but used for the wrong purposes.

Even better, establish a family safe word or passphrase today. Choose something simple that everyone can remember but that would never appear online.

If you ever receive one of these calls, ask for the safe word. If they can't provide it, assume it's a scam until you can verify the situation by calling your child directly or contacting another trusted family member.

As AI continues to improve, these scams will only become more convincing. The same technology powering innovation is also giving criminals new tools to exploit unsuspecting families. Stay alert. Verify before you trust. A few extra minutes could save you thousands of dollars and a great deal of heartache.

Is It Boom or Bust for Micron?

It is hard to argue with Micron's incredible stock performance. Through July 2, the shares were up 242% year to date and an astonishing 701% over the previous 12 months. Even after recently falling about 22% from their peak, investors are still debating whether the company has much more room to run.

The good news is that Micron has locked in 15 new customers under long-term supply agreements, with some contracts extending as long as five years. Many of these agreements include customer deposits, giving the company excellent revenue visibility and reducing uncertainty over future sales. For investors, that is exactly the kind of stability they like to see.

But every smart investor should also ask: What is the downside?

While those contracts provide a strong foundation, they do not guarantee that demand will remain as strong over the long term. Unless a customer goes bankrupt, the contracts are largely locked in, but technology changes quickly. High prices and limited supply often encourage innovation, and the AI memory market is no exception.

Several companies are developing new architectures that reduce or even eliminate the need for high-bandwidth memory (HBM), which has been one of Micron's biggest growth drivers. As companies search for lower-cost and more efficient alternatives, demand for HBM could eventually soften.

Nvidia also signaled in June that it is redesigning portions of its upcoming Vera Rubin AI platform to use memory more efficiently. While Nvidia remains a major customer for HBM, improvements in memory efficiency could reduce the amount of HBM required per AI system over time.

Meanwhile, newly public chipmaker Cerebras has taken an entirely different approach. CEO Andrew Feldman has said the company's wafer-scale AI chips do not use HBM at all, arguing that it is too expensive and supply constrained. If other AI hardware companies pursue similar designs, it could create additional competition for HBM.

None of this means Micron's growth story is over. The company's long-term contracts provide meaningful protection, and AI demand remains exceptionally strong today. However, investors should remember that today's shortages and premium pricing often inspire tomorrow's technological breakthroughs.

The question for Micron investors is whether HBM remains the industry standard for years to come or whether innovation eventually reduces the need for it. If demand for HBM begins to slow, Micron's remarkable growth could also begin to moderate.