Airplanes are expensive, but banks are comfortable lending money against them. An aircraft can remain useful for decades, there is an established global resale market, and lenders have years of historical data showing what different models are likely to be worth over time. Nvidia now wants Wall Street to start thinking about its AI chips in a similar way. The company is working to turn GPUs—the hardware powering the artificial-intelligence boom—into financeable assets that can be used as collateral for enormous infrastructure loans.

The ambition is massive. Nvidia is supporting a financing initiative that could eventually reach roughly $500 billion, helping AI companies and cloud providers fund the servers and data centers needed to run increasingly powerful models. The basic idea is familiar: instead of requiring a company to pay billions upfront for computing infrastructure, lenders finance the purchase and use the equipment itself as part of the collateral. Nvidia argues that its GPUs are productive, transferable assets that can continue generating revenue for years.

Wall Street’s problem is that an Nvidia GPU is not an airplane. Commercial aircraft can remain in service for 20 or 30 years. AI hardware evolves dramatically faster. A chip that is considered cutting edge today may face substantially more powerful competition only a few years from now. Banks therefore tend to depreciate AI chips over roughly three or four years, even as Nvidia argues that some of its hardware may remain useful for as long as a decade. That difference in assumptions matters enormously when billions of dollars are being lent against the equipment.

The issue comes down to residual value. If a borrower stops making payments on an aircraft loan, the lender can repossess the plane and potentially lease or sell it to another airline. The market for used aircraft is well established. If a cloud company defaults on a loan backed by thousands of AI chips, the bank needs confidence that another customer will actually want those chips—and that they will still be worth enough to recover the outstanding debt. That secondary market is much younger, and technology changes far faster than aviation.

This is why lenders increasingly want more than the hardware itself. Banks and credit investors are looking for long-term customer contracts, predictable revenue streams and guarantees from financially strong companies before providing large loans. A rack full of expensive GPUs may look valuable, but the real security could ultimately be the contract from a major customer promising to pay for the computing capacity those GPUs produce. In finance, the ability of an asset to generate reliable cash flow can matter more than its original purchase price.

If Nvidia succeeds, however, the implications could be enormous. Turning GPUs into a recognized financial asset would allow smaller AI infrastructure companies to raise far more capital without relying entirely on equity investors. Airlines do not pay cash for every new aircraft because specialized financing markets allow them to spread the cost across years of operation. A similar system for computing could help fund the trillions of dollars of infrastructure expected to be required for AI.

It would also represent an important evolution in Nvidia’s business model. The company would no longer simply manufacture chips and sell them to customers. It would help build the financial system that allows customers to afford those chips. Nvidia has already used investments, guarantees and financing relationships to support companies throughout the AI ecosystem. Building a mature lending market around GPUs would take that strategy further by making access to capital another tool for expanding demand.

There is a risk in that strategy. When sellers begin helping finance buyers, it becomes harder to distinguish pure customer demand from demand supported by financing. The same dynamic has appeared in industries ranging from automobiles and heavy machinery to telecommunications equipment. Financing can accelerate adoption and create enormous markets, but it also increases financial connections between manufacturers, lenders and customers. If the economics of AI infrastructure weaken, those connections could transmit losses through the system.

The bigger lesson is that an asset does not become good collateral simply because it is expensive. Banks care about durability, resale markets, predictable depreciation and the ability to recover money if something goes wrong. Nvidia is trying to convince Wall Street that computing power deserves to join aircraft, real estate and industrial equipment as a major financeable asset class. Whether that happens may depend less on how powerful the chips are today and more on a much harder question: what will those chips still be worth five years from now?

The next phase of business AI is moving beyond chatbots.

Meta is now putting its AI agent, Muse, directly into the workflows of small businesses, allowing owners to connect the agent to many of the applications they already use to run their companies. The announcement comes as Meta simultaneously builds a dedicated enterprise AI business, signaling a broader shift from AI that answers questions to AI that can work across a company’s operations.

For small-business owners, that distinction could matter considerably.

Muse Can Now Connect to the Business Behind the Business

Meta’s new Muse for Small Business can connect to Facebook and Instagram business accounts as well as services including Shopify, QuickBooks, Slack, Canva, Dropbox, Asana, Notion, Stripe, Klaviyo, Zoom and HighLevel.

The idea is relatively simple: instead of asking an AI about information that has been manually copied into a conversation, the business can give the agent access to the systems where its information already lives.

Meta says an owner can give Muse goals such as analyzing sales performance, finding new customers, improving advertising or figuring out which business expenses deserve attention. Muse can then work across the connected information to produce recommendations and carry out portions of the work.

That represents an important change in the AI market.

For years, the typical small-business AI pitch was essentially:

“Ask our chatbot a question.”

The emerging pitch is:

“Give our agent a business objective.”

The Approval Button May Be More Important Than the AI

There is an important limitation built into Meta’s system.

Meta says Muse will not publish content, send messages or spend money without the user’s approval.

That distinction becomes increasingly important as AI agents gain access to business systems.

An AI that drafts a Facebook advertisement is one thing.

An AI that can access a company’s advertising account, customer information, financial records and payment systems is something entirely different.

The more useful agents become, the more consequential their mistakes can become. That is one reason the industry is increasingly focused not only on what AI agents can do, but on what they are permitted to do without human authorization.

Small Businesses Are Becoming the Next AI Battleground

Meta’s announcement did not happen in isolation.

On September 28, Meta launched Meta Enterprise Platform, a new business focused on bringing its AI models, agents and developer tools to companies. Meta specifically identified Muse, Meta Business Agent, Muse API and Muse Code as components of the platform.

A day later, OpenAI announced Dots, always-on AI agents designed to continue working on goals between conversations. OpenAI says a Dot can operate from its own cloud computer, connect to the applications it needs and continue making progress for the user.

The timing is revealing.

The major AI companies are increasingly competing for the same thing: the work performed inside businesses.

That potentially puts traditional business software in an interesting position.

If an AI agent can sit between a business owner and dozens of applications, the owner may eventually interact less with individual software products and more with the AI layer connecting them.

What This Means for a Local Business

Consider a small plumbing company.

Today, the owner might have one system for leads, another for scheduling, another for accounting, another for marketing, another for customer communication and another for social media.

The owner or employees have to move information between those systems.

An agent changes the potential workflow.

A business owner could theoretically ask:

«“Find out where our leads are coming from, identify the campaigns producing the best customers, follow up with prospects who haven’t booked, and prepare next week’s marketing plan.”»

The important development is not that AI can write the resulting report.

It is that the agent can increasingly access the underlying business information required to produce it.

Meta is explicitly building toward this model by connecting Muse to the applications small businesses already use.

The Opportunity for Businesses Is Also a Warning for AI Service Providers

This shift creates a more complicated market for companies selling AI automation.

An AI receptionist that simply answers questions may no longer be the entire opportunity.

The larger opportunity is connecting AI to the business’s actual workflow: lead generation, qualification, follow-up, scheduling, customer records, payments, marketing and reporting.

At the same time, some basic automation services could become commodities as platforms such as Meta and OpenAI incorporate more capabilities directly into their products.

That does not necessarily eliminate the need for implementation.

For many small businesses, the difficult part isn’t having access to an AI agent. It is determining what the agent should actually do, which systems it should access, what permissions it should have, and where humans must remain in control.

That creates a different kind of service opportunity: designing and managing the business process around the AI rather than simply selling access to the AI itself.

The Bigger Story

The significance of Meta’s announcement is therefore bigger than Muse.

Meta is betting that small businesses will eventually want AI that understands their business context and can operate across the software stack rather than another isolated chatbot.

OpenAI is making a similar bet with its always-on Dots.

The competitive question is increasingly shifting from:

“Which AI gives the best answer?”

to:

“Which AI can actually get the work done?”

For small businesses, that could ultimately be a much more consequential question.

The companies that figure out how to safely delegate repetitive operational work to AI agents could gain significant leverage without necessarily adding more employees.

But the businesses adopting these systems will also have to treat permissions, oversight and security as part of the implementation—not as an afterthought.

The AI assistant is becoming an AI operator.

And for small businesses, that may be the development worth watching most closely.

For decades, businesses have charged different prices in different places. A meal in Manhattan costs more than the same meal in a small town because rent, wages, competition and customer demand are different. McDonald’s is now bringing much more technology into that decision. The company uses a machine-learning pricing engine that analyzes millions of transactions across its nearly 14,000 U.S. restaurants and recommends what individual menu items should cost at specific locations. Among the factors the system considers are nearby competitors and an estimate of how much customers in that area are willing to pay.

That last part changes the conversation. Pricing has traditionally started with a relatively simple question: what does this product cost us to sell, and what margin do we need to earn? Modern data allows companies to ask something much more powerful: what is the highest price this particular market will accept before customers start walking away? McDonald’s system can evaluate local purchasing behavior and compare publicly available prices at nearby competitors such as Burger King and Wendy’s to help recommend what it calls an “optimal price.”

The result can be surprisingly different prices even within the same city. Reuters found one company-operated McDonald’s in Fresno, California, selling a Big Mac for $5.69 while another company-operated location just two miles away charged $6.89—a difference of about 21%. Reuters could not determine whether that specific difference came from the AI system or other local factors, but franchisees told the news organization that the pricing engine has widened some existing price differences between nearby restaurants.

The most attention-grabbing example dates back to 2023, when McDonald’s pricing tools recommended that a Connecticut franchisee charge roughly $18 for a Big Mac meal at a restaurant located off a state turnpike. That does not mean McDonald’s now universally charges $18 for Big Mac meals, and the company says franchisees remain free to set their own prices. But several franchisees told Reuters they felt pressure to follow the company’s recommendations, with corporate representatives sometimes contacting operators whose pricing differed from the suggested levels.

This is where smart pricing starts becoming a much bigger business question. Companies have always tried to understand what customers are willing to pay. Airlines change fares based on routes and demand. Hotels charge more during busy weekends. Gas stations across the street from one another constantly react to competitors. What AI changes is the precision. Instead of managers occasionally reviewing prices, machine-learning systems can analyze enormous amounts of information and identify pricing opportunities humans may never notice.

That can be extremely valuable. Even a small improvement in average pricing multiplied across billions of customer transactions can translate into enormous amounts of additional revenue. But there is another side to that calculation. Customers do not experience pricing as a spreadsheet optimization problem. They experience it emotionally. A customer who discovers that the same Big Mac costs significantly more a few miles away may not think about rent, demand elasticity or machine-learning models. They may simply believe the company is charging whatever it thinks it can get away with.

That makes perceived fairness increasingly important. A technically perfect pricing algorithm could still damage a brand if customers believe the resulting prices are unreasonable. McDonald’s is particularly exposed to this tension because value has historically been a major part of its identity. The company is simultaneously trying to win back price-sensitive customers and has acknowledged that persistent inflation is weighing on restaurant traffic. A pricing system that maximizes individual menu-item economics therefore has to coexist with a broader brand promise that customers can still afford to eat there.

There are regulatory questions as well. Courts and regulators are increasingly examining whether algorithmic pricing systems can facilitate improper coordination when businesses that are technically competitors use common pricing technology. McDonald’s itself warns franchisees using the pricing portal about potential antitrust scrutiny because individual restaurant owners can be considered competitors. The issue is not simply whether AI raises prices. It is whether increasingly centralized pricing recommendations begin influencing supposedly independent businesses in ways regulators consider problematic.

The broader lesson reaches far beyond fast food. Retailers, hotels, airlines, entertainment companies and online marketplaces increasingly have access to extraordinary amounts of customer data. AI gives them the ability to turn that information into increasingly precise estimates of willingness to pay. Economically, that is powerful. Psychologically, it can be dangerous. The better companies become at calculating exactly how much a customer will tolerate, the more customers may begin wondering whether the company is offering them a fair price or simply extracting the maximum possible amount.

AI may ultimately make pricing dramatically smarter. But businesses will still have to answer a very human question that an algorithm cannot solve for them: just because a customer is willing to pay more, does that mean you should charge them more? The companies that use AI pricing successfully may not be the ones that squeeze every possible dollar from each transaction. They may be the ones that optimize revenue without destroying the customer’s belief that the deal is still fair.

When consumers think about what makes a car expensive, they usually picture steel, electronics, batteries, factory workers and transportation. But one of the largest costs built into a modern vehicle can come from something the customer never physically sees: regulation. The U.S. Transportation Department estimates that newly finalized fuel-economy standards could reduce General Motors’ technology costs by about $20.4 billion through 2031. Across the auto industry, the government estimates manufacturers could avoid roughly $60.6 billion in technology costs, equivalent to about $1,289 per vehicle.

Those numbers reveal how deeply regulation can shape the design of a product before it ever reaches a showroom. Under the previous fuel-economy standards, automakers would have needed to invest more heavily in technologies that reduce fuel consumption, including more efficient engines, hybrid systems, lighter materials and a greater share of electric vehicles. The new rules lower the required fleetwide fuel economy target, giving manufacturers more flexibility over which technologies they use and which vehicles they produce.

For GM, the difference is enormous. The previous rules were estimated to require about $31.7 billion in technology spending through 2031. Lowering those requirements does not suddenly make factories cheaper or reduce the price of steel. Instead, it changes how much engineering manufacturers must put into making their vehicles satisfy government standards. That is an important distinction because engineering requirements can influence nearly every part of a vehicle, from the engine and transmission to materials, software and the mix of models a company chooses to sell.

The new rules call for a fleetwide average of roughly 34.9 miles per gallon by 2031, compared with about 50.4 mpg under the previous standards. The government estimates that the change will reduce manufacturers’ compliance costs, although it also projects that drivers will consume more gasoline and ultimately spend more on fuel over the life of their vehicles. That illustrates another important reality of regulation: lowering costs for the manufacturer does not necessarily eliminate the cost. Sometimes it shifts where the cost appears.

The auto industry is particularly sensitive to this because manufacturers plan products years before customers ever see them. A regulation scheduled for 2030 can influence vehicles engineers are designing today. Factories may need to be retooled. Suppliers may need new contracts. Billions of dollars can be committed to battery plants, engines, transmissions or electric-vehicle platforms based partly on what companies believe future rules will require. When those rules change, entire investment plans can change with them.

This is why regulation should be viewed as part of a product’s economics rather than something sitting outside the business. A manufacturer does not simply calculate the cost of materials and labor and then add profit. It also has to calculate what the product must do to legally enter the market. Safety standards, emissions rules, fuel-efficiency requirements, labeling laws and testing procedures can all influence how much the final product costs to develop and manufacture.

The same principle exists far beyond automobiles. Pharmaceutical companies spend enormous amounts complying with testing and approval requirements. Banks build expensive compliance departments because of financial regulation. Food companies redesign packaging when labeling laws change. Technology companies alter products to comply with privacy rules. Construction firms build around safety and environmental standards. In every case, regulation becomes another input in the cost of producing the product.

That does not mean regulation is simply wasted expense. Rules frequently exist to produce benefits that markets may not create on their own, including safer products, lower pollution or more transparent information. The important business lesson is that regulatory decisions have economic consequences regardless of whether someone believes a particular rule is good or bad. Change the requirement and companies change where they spend money.

There is another strategic consequence as well. Companies that anticipate regulatory changes correctly can gain an advantage over competitors that invest heavily in the wrong direction. If one automaker commits billions to a technology based on a rule that later disappears while another retains more flexibility, their cost structures can diverge dramatically. Regulation therefore becomes part of competitive strategy, capital allocation and long-term planning.

GM’s projected $20.4 billion reduction is a dramatic example of something consumers rarely see. A large portion of what determines the cost of a product can be decided long before materials arrive at the factory. Governments set requirements. Engineers design around them. Companies invest billions to comply. Only then does the customer see the finished vehicle and its price tag. The cost of making a product is not determined only by what goes into it. It is also determined by what governments require that product to become.

For years, the technology industry has operated with a simple assumption: consumers always want more. More features, more connectivity, more automation, more convenience. But a growing consumer trend suggests the opposite may also be true. As artificial intelligence becomes more powerful and more integrated into daily life, many people are becoming more interested in products and experiences that do less, not more. Interest in “analog hobbies” has surged dramatically, along with growing attention to “offline living,” as consumers increasingly seek activities and products that feel tactile, slower and less mediated by screens.

That shift matters because it suggests that simplicity itself may be becoming a premium feature. A product does not always become more appealing by doing more things. Sometimes it becomes more valuable by doing fewer things and doing them in a calmer, more focused way. That is helping create opportunities for businesses built around film cameras, notebooks, knitting, watercolor painting, vinyl records, phone-free gatherings, simpler devices and other low-tech or deliberately limited experiences.

The broader appeal is easy to understand. Digital life has become intensely crowded. Phones are not just communication tools anymore. They are work devices, entertainment hubs, shopping platforms, social feeds, cameras, payment systems and constant sources of interruption. AI may make that environment even more powerful, but also more overwhelming. When technology becomes too present, too fast or too demanding, consumers often begin looking for products and environments that give them back a sense of control.

That is why the analog trend is not simply nostalgia. It is also about boundaries. People are not necessarily rejecting technology altogether. Most still rely on it for navigation, communication, work and convenience. What they seem to be seeking is a better balance. A record player, a handwritten notebook or a knitting project offers something digital life often does not: a single purpose, tactile engagement and an experience that does not constantly compete for attention. In that sense, analog products are not just old-fashioned. They are increasingly becoming tools for managing overstimulation.

This shift is creating interesting business opportunities. Small stores built around vintage media, retro technology, crafts and in-person gatherings are finding demand from customers who want less screen time and more physical interaction. Larger retailers are noticing the same pattern. Simpler phones, retro collectibles, paper like writing devices, vinyl records and hobby-related goods all fit into a market where consumers are looking for products that feel more intentional. What once might have looked like niche taste is beginning to resemble a real retail category.

There is a larger business lesson here. Companies often assume innovation means adding functionality. But sometimes the smarter move is subtractive. Businesses can create value by reducing friction, reducing distraction and narrowing a product’s purpose. A device that helps you read and take notes without turning into a social-media portal can be more appealing than a more powerful but more distracting alternative. A hobby that requires patience and physical effort can feel more satisfying precisely because it is not optimized for speed.

The analog trend also highlights an important emotional dimension of consumer behavior. People do not buy products only for efficiency. They buy for identity, ritual, comfort and how a product makes them feel. Technology can solve practical problems while still leaving people hungry for texture, slowness and physical presence. A film camera is less convenient than a smartphone camera. A handwritten planner is slower than an app. A vinyl record is less efficient than streaming. But that is often exactly the point. The imperfection and limitation are part of the appeal.

This creates a powerful form of differentiation. When everything becomes digital, being intentionally analog can stand out. A business offering face-to-face community, tactile experiences or simpler tools may begin to feel more distinctive as AI expands further into everyday life. The more software automates, predicts and optimizes, the more some consumers may value things that feel human, manual and imperfect.

AI is making technology more capable than ever. But that does not automatically mean consumers want every part of life to become more digital. In many cases, the opposite may happen. The more connected and automated the world becomes, the more attractive simpler products and offline experiences may appear. The next successful businesses may not only be the ones building the smartest technology. They may also be the ones giving people a reason to put that technology down.