Most assets simply can’t reinvent themselves every few quarters, but Ethereum (ETH +4.21%) arguably does just that. After pushing two major upgrades, Pectra and Fusaka, in 2025, the chain has another two big improvements on the docket for 2026.

Nonetheless, the coin’s price is down by 38% during the past three months alone, largely for macro reasons that are well beyond its control. Thus it’s likely undervalued, and potentially by quite a lot. Does that make it a screaming buy with a hearty investment of $5,000?

The upgrade pipeline is solid, but it can’t guarantee returns

Ethereum’s 2025 upgrades were a lot more than cosmetic improvements, and they laid the technical groundwork for a lot of the follow-on work that’s going to happen this year. This stuff might sound boring (and it might actually be) but knowing what’s going on with it is key to appreciating the chain’s place in the crypto sector’s competitive landscape, not to mention its future opportunities for growth.

The Pectra upgrade went live in May 2025, and it bundled changes aimed at providing better wallet UX, more efficient staking, and more throughput for Layer-2 (L2) chains. Fusaka followed on Dec. 3, and its headline feature, peer-to-peer data availability sampling (PeerDAS) is also a game changer for the chain’s ability to provide rapid performance at scale, and substantially cheaper than before. Today, the chain’s average transaction fees are roughly 75% lower than three years ago, with an average token swap now costing about $0.30, so these successive upgrades are definitely succeeding in making Ethereum a cheaper and easier technology to use.

Today’s Change

(4.21%) $85.83

Current Price

$2123.98

Key Data Points

Market Cap

Advertisement

$256B

Day’s Range

$2014.22 – $2136.13

52wk Range

$1398.62 – $4946.05

Advertisement

Volume

32B

For 2026, the next upgrade, Glamsterdam, will build on those past successes while also adding new censorship resistance features. But, if the coin’s price performance after past updates is any indication, investors simply can’t count on a boost.

There’s no rush to buy it

There’s not exactly a rush to buy Ethereum before Glamsterdam drops.

Ethereum’s upside comes from being the settlement layer that L2s and on-chain finance route through. Given that its upgrades tend to reduce transaction costs rather than increase them, the coin’s value capture from the traffic it supports is still very weak, and it would likely take a deluge of new traffic to move the needle for investors. Realistically, the new traffic will probably ramp up slowly over time, assuming it arrives at all, so buying the coin means getting exposure both to the value generated from the improvement of its underlying tech and also the value generated from people using it to pay for decentralized finance (DeFi) apps and services.

But it’s still very much an asset worth owning, as it’s one of the most important in the crypto sector. An investment of $5,000 buys roughly 2.5 coins, which is enough exposure in case 2026’s development road map plays out such that the coin’s price significantly rises, which is still possible.

Of course, if you’re usually intolerant of risk, it’s probably better to aim for a much smaller allocation.

Artificial intelligence is starting to do things that were formerly the exclusive domain of humans, including tasks like holding and moving money. If the “agentic AI” trend sticks, it’s thus reasonable to assume that more financial activity will be initiated by software, and, perhaps even for the benefit of that software rather than for the benefit of humans.

That brings up a fun, slightly unsettling question for investors: Could Bitcoin (BTC 1.03%) benefit by becoming a preferred store of value for AI agents?

What AI agents will actually optimize for

In practice, the AI agents of today don’t have any need for money in the sense that a human might. They’re machines designed to identify market patterns, assist with payment routing, manage liquidity in key accounts, and monitor fraud risk.

That set of jobs implies handling a very particular kind of money. In short, for an AI agent to excel at those tasks, it needs to operate within a system with low, stable costs and clear integration points for basic functionalities like identity verification and trade authorization. If those requirements aren’t met, the agent can’t do much of anything because the company or individual running it will be loath to eat the operational costs and regulatory risks associated with letting it continue, even if it’s possible to do so.

Today’s Change

(-1.03%) $-721.23

Current Price

$69220.00

Key Data Points

Market Cap

Advertisement

$1.4T

Day’s Range

$67656.00 – $71487.00

52wk Range

$60255.56 – $126079.89

Advertisement

Volume

67B

So even if AI agents become a real theme in the world of managing investments and making trades — and they probably will — the initial wave of agent activity will probably concentrate in quite narrow and controlled workflows rather than a sudden, industrywide automation of everything. And there simply aren’t many ways for AI to change or improve upon the Bitcoin mining process either.

Therefore, we should not expect AI agents to immediately cause noticeable changes in Bitcoin’s price, as they might not.

Where Bitcoin could see upside

The best case for Bitcoin here is not that it becomes a spendable asset for agents. It’s simply a bad fit for that purpose; it’s slow and expensive to use, and it lacks any smart contract infrastructure for automated systems to hook into gracefully. Nonetheless, Bitcoin could still gain a lot from the rise of AI if it becomes the reserve store of value that agents use to invest their earnings, assuming they ever have any.

It’s a decent choice for that purpose because it has a fixed supply schedule and a governance culture that makes major changes slow and contentious, both of which are good features for those seeking a long-lived store of value that doesn’t require a human to handle. Of course, there are other cryptocurrencies that could fill that same role, though none are as widely trusted as Bitcoin.

So, what should investors watch for if they want to see whether the AI upside in Bitcoin is actually going to play out as described here?

Look for financial institutions building agent-ready Bitcoin custody solutions with policy controls, and for large financial businesses explicitly describing Bitcoin as a strategic reserve asset inside their AI-driven operations.

Until those hints appear, it’s a lot more reasonable to treat AI as a modest tailwind for Bitcoin.