Reading Your AI Startup Offer: Vesting, Exercise and Dilution

What you need to know

• The headline number is the least important number. "0.5% equity" or "20,000 options" means nothing until you know the vesting schedule, the strike price, the fully diluted share count and what happens to it after the next funding round.
• Vesting is still mostly 4 years with a 1-year cliff. That remains the industry default, though the AI hiring market has produced more variation — shorter cliffs, front-loaded vesting, refresher grants — than past cycles.
• The exercise window can quietly erase your equity. Many companies still give you only 90 days after leaving to buy vested options before they expire — a real cost that catches people off guard. A growing minority offer far longer windows.
• Dilution is not a bug — but you should model it. Every funding round shrinks your ownership percentage unless you have pro-rata rights. What matters is whether the pie is growing faster than your slice is shrinking.
• Tax treatment genuinely differs across India, the UK and the US. ISOs, NSOs, UK EMI options and India ESOPs are taxed at different points and different rates. Do not assume what applied to a friend's offer elsewhere applies to yours.

Start with the vesting schedule, not the headline number

Vesting spreads your equity grant out over time so you earn it by staying, rather than owning it all on day one. The schedule that has dominated venture-backed startups since the late 1990s — codified by standard Silicon Valley legal templates and now built into every cap-table tool — is four years total with a one-year cliff: nothing vests for your first twelve months, then 25% vests on your one-year anniversary, and the remaining 75% vests monthly over the following three years.

The cliff is the part people underestimate. It is a hard binary line: leave on day 364 and you walk away with zero equity from that grant, no matter how close you were. Leave on day 366 and you keep the quarter that had already crystallised — deliberately, to discourage short stays and give the company a clean exit from a hire that clearly is not working out.

The competitive AI hiring market has introduced more variation on top of this default than most previous cycles. Some companies now offer six-month cliffs, front-load year-one vesting to compete for scarce specialists, or issue refresher grants well before the original four years are up. None of this is universal, so ask for the actual vesting schedule in writing rather than assuming the default applies.

The exercise window: the clause that quietly cancels equity

Vesting only tells you what you have earned the right to buy. Owning stock options is not the same as owning shares — an option is the right to purchase shares later, at a fixed strike price (usually the fair market value on your grant date, set by an independent valuation). To actually own the shares you must exercise the option: pay the strike price, in cash, for the shares you want. That right to exercise does not last forever, especially after you leave.

The historical default — still the most common arrangement — gives departing employees just 90 days after their last day to exercise any vested options before they expire worthless. Ninety days also happens to be the maximum window during which a US Incentive Stock Option can be exercised after termination and keep its favourable ISO tax treatment; exercise later and the option is automatically reclassified as a Non-Qualified Stock Option, even if the plan technically allows longer. For someone who has vested a meaningful stake over several years, finding real cash within three months of leaving — often while also owing tax on the exercise itself — is a genuine barrier, and it is one reason many people leave equity on the table when they change jobs.

Since around 2014, a minority of companies have pushed back on this default by extending the Post-Termination Exercise Period, or PTEP. Quora is widely credited as one of the first to move to a ten-year window, back in 2014; Pinterest followed with a seven-year window for staff with at least two years' service; Asana, Coinbase, Palantir and Square adopted longer windows in the years after. Extended PTEPs remain an established but still uncommon retention signal rather than the norm. Even with an extended PTEP, the tax mechanics usually still convert the option from ISO to NSO status once the 90-day IRS window passes — a longer window buys time to raise cash and decide, but does not avoid that conversion.

Double-trigger acceleration: the clause that protects you in a sale

Vesting assumes you stay long enough to earn your equity organically. What happens if the company is acquired before you are fully vested? This is where acceleration clauses matter — and the one worth specifically asking about is double-trigger acceleration.

Double-trigger acceleration vests some or all of your remaining unvested equity only when two separate events both occur: first, the company is acquired or undergoes a change of control; and second, within a defined window around that event (commonly a matter of months), you are terminated without cause or resign for a legitimate reason such as a significant cut to your pay or responsibilities. If the company is sold and you keep your job on similar terms, nothing accelerates — you carry on vesting normally. If you are let go as part of the integration, your unvested equity accelerates instead of simply being cancelled.

This is deliberately different from "single-trigger" acceleration, which vests equity purely on the change of control itself, regardless of what happens to your job afterwards. Single-trigger sounds better on paper but is far less common in practice — it can make a company less attractive to acquirers, who inherit a workforce with no more vesting to earn. Double-trigger is the market-standard compromise, increasingly common at Series B and later. Earlier-stage startups may not offer it at all — a fair, specific question to raise before you sign.

Dilution: why your percentage shrinks even as the company grows

Every time a startup raises a new funding round, it issues new shares to the investors writing the cheque. Unless you have a pro-rata right to buy into that round yourself — rare for employees, mostly reserved for larger existing investors — your percentage ownership goes down even though your share count stays the same. This is dilution, and it is not a sign anything has gone wrong; it is the normal mechanical consequence of raising the capital a company needs to grow.

Three mechanics matter most for reading an offer:

• Pre-money vs post-money valuation. If a company is valued at $40M pre-money and raises $10M, it is worth $50M post-money, and new investors own roughly 20%. Everyone who held shares before — founders, investors, option holders — is diluted by roughly that proportion, before any option pool changes.
• The option pool shuffle. Investors typically require the company to top up an option pool for future hires as part of the round, usually carved out of the pre-money valuation — so existing shareholders, not new investors, absorb most of that dilution. A round often dilutes option holders by more than the headline investment percentage suggests.
• Fully diluted share count. Any percentage in your offer should be expressed against the fully diluted count — all shares, unvested options, unallocated pool and convertible instruments counted as issued. A percentage quoted against a narrower count will look bigger than your real ownership.

Funding stage	Illustrative valuation	Your 0.20% grant, fully diluted	What just happened
At your grant (Series A)	$40M post-money	0.20% → ~$80,000 notional value	Starting point — notional only, nothing is liquid
After Series B	$120M post-money	~0.15% → ~$180,000 notional value	Diluted by ~25%, but company value tripled — your slice is smaller but worth more
After Series C + pool top-up	$300M post-money	~0.11% → ~$330,000 notional value	Diluted further by the round and a new option pool for scaling hiring

Fully hypothetical numbers for illustration only — not a forecast, and most startups never reach a liquidity event at all. Real dilution per round depends on round size, pool top-up size and your specific cap table.

Dilution and value creation are not opposing forces — they can both happen at once, and the outcome depends on whether the company's value is growing faster than your percentage is shrinking. A shrinking slice of a fast-growing pie can still be worth more in absolute terms; a shrinking slice of a stagnant or down-round company is a genuine loss. That is why "down rounds" — raising at a lower valuation than the previous round — are especially painful: dilution happens without the offsetting value growth. This is also the arithmetic behind the cash-versus-equity trade-off we cover in startup vs Big Tech vs AI lab: your AI engineering path — a bigger slice of an early, unproven company is not automatically worth more than a smaller slice of one that is already scaling.

ISO vs NSO vs UK EMI vs India ESOP: how the structures differ

The mechanics above apply broadly across markets. What differs sharply is how each structure is taxed — conflating the US, UK and India systems is one of the most costly mistakes builders make comparing offers across borders. High-level comparison below; treat every rate as illustrative and confirm current figures with a qualified adviser.

Structure	Who can receive it	Tax at grant	Tax at exercise	Tax at sale
ISO (US — Incentive Stock Option)	Employees only	None	None for regular tax; the spread can trigger Alternative Minimum Tax	Long-term capital gains rates if held 1yr+ from exercise and 2yr+ from grant; otherwise a "disqualifying disposition" taxed partly as ordinary income
NSO (US — Non-Qualified Stock Option)	Employees, contractors, advisors, directors	None (if priced at fair market value)	Ordinary income tax on the spread between strike price and fair market value	Capital gains tax on any further appreciation since exercise
UK EMI option	Employees of qualifying small/mid-size UK trading companies, working 25hrs/week or 75%+ of their time for the company	None, if granted at unrestricted market value	None, if exercised at the market value set at grant	Capital Gains Tax on sale; Business Asset Disposal Relief can apply a reduced rate if the option was held long enough
India ESOP	Employees (structure varies by company; DPIIT-recognised startups have some deferral relief)	None	Taxed as a "perquisite" — treated like salary income at your slab rate, on the spread between fair market value and exercise price	Capital gains tax on any further appreciation since exercise, with the exercise-date fair value used as the cost base

High-level comparison only, as of July 2026 — not exhaustive and not tax advice. Structures, eligibility rules and rates change; always confirm current treatment with a professional in the relevant jurisdiction.

United States: ISO vs NSO

The core US distinction is timing. An NSO is taxed as ordinary income on the "spread" — the gap between what you pay to exercise and the shares' fair market value at that moment — the instant you exercise, whether or not you sell. An ISO generally owes no regular income tax at exercise; the gain is deferred until sale, and if you hold long enough after both exercise and grant, the entire gain can qualify for long-term capital gains rates, typically lower than ordinary income rates. The catch: the ISO spread at exercise can still trigger the Alternative Minimum Tax, a parallel US tax calculation that can create a real cash bill with no sale proceeds to pay it from — one of the most common expensive surprises in US equity compensation, worth modelling with a tax professional before a large ISO exercise. Only employees can receive ISOs; contractors, advisors and non-employee directors can only receive NSOs, and a $100,000-per-year cap (grant-date value) limits how much ISO value can vest favourably in any calendar year — the rest is automatically treated as an NSO.

United Kingdom: EMI options

The UK's Enterprise Management Incentive scheme is the tax-advantaged route most UK startups use, and it is generous when the qualifying conditions are met: under rules taking effect from April 2026, the granting company must run a qualifying trade, have gross assets under £120 million, and employ fewer than 500 full-time-equivalent staff; an individual can receive up to £250,000 of EMI options in any three-year period, up to £6 million in total across the scheme. Employees must work at least 25 hours a week, or 75%+ of their working time, for the company. Granted at unrestricted market value, EMI options generally carry no income tax or National Insurance at grant or exercise — a meaningful advantage over an unapproved option. Capital Gains Tax applies on eventual sale; as of July 2026, the standard CGT rate for higher and additional rate taxpayers on shares is 24% (in place since late 2024), and Business Asset Disposal Relief can reduce this to 18% from 6 April 2026 (up from 14% the prior tax year) on the first £1 million of qualifying gains, subject to a minimum holding period. These figures change most tax years — verify current rates on gov.uk or with a UK adviser before relying on them.

India: ESOP perquisite and capital gains tax

India's ESOP taxation, per the Income Tax Department's own published guidance, works in two stages. First, at exercise, the difference between the fair market value on the exercise date and the price you paid is taxed as a "perquisite" — treated like salary income, at your applicable slab rate, with your employer typically deducting tax at source. Second, when you eventually sell, any further gain above that exercise-date fair value is taxed separately as a capital gain — long-term or short-term depending on holding period (broadly, over 24 months for unlisted shares counts as long-term). As of July 2026, tax-filing platforms report long-term capital gains on unlisted shares at a flat rate without indexation, though the precise figure should be confirmed against current guidance, since Indian capital gains rules have changed more than once in recent years. Employees of DPIIT-recognised eligible startups have historically had access to a deferral of when this tax falls due — worth asking your employer's finance team about directly.

What your equity might actually be worth: a worked example

None of this is a forecast — most startup equity is worth substantially less than the number in an offer letter suggests, and a large share of startups never reach a liquidity event at all, in which case options are worth nothing however good the vesting schedule looked. With that in mind, here is the rough arithmetic worth doing on any offer before weighing it against a cash-heavier alternative.

Illustrative scenario. Suppose you are offered 0.30% of a company, fully diluted, at a $30M post-money valuation, with a standard 4-year vesting schedule and a 1-year cliff, and a strike price set at the current $0.40-per-share fair market value.

1. Notional value at grant: 0.30% of $30M is $90,000 — not cash, not guaranteed, and will be diluted by every future round. A starting reference point, not a number you can spend.
2. Adjust for dilution over time. If two more rounds each dilute existing holders by roughly 15-20% (illustrative, not a rule), your 0.30% might realistically be closer to 0.20% by any exit — even as the company's value grows.
3. Apply a realistic probability of any payout at all. Venture-backed startup outcomes follow a well-known power-law pattern: most return little or nothing to option holders, while a small handful produce most of the aggregate return. Treating your equity as having meaningfully less than a coin-flip chance of paying out is more honest than treating it as likely.
4. Subtract exercise cost and tax due. If you leave before an exit, you may need real cash to exercise within your PTEP window, and depending on structure and jurisdiction, may owe tax at exercise before receiving any cash back.
5. Compare the risk-adjusted figure to the cash you'd give up. The honest comparison is guaranteed cash today versus a small, uncertain, multi-year, tax-encumbered probability of a future payout — not "$90,000 versus a smaller salary".

Red flags to watch for in an offer letter

• No mention of the fully diluted share count. A percentage or share count with nothing to divide it by tells you almost nothing about your real ownership.
• An exercise window shorter than the industry default with no explanation. Ninety days is already tight; anything shorter, or vague wording that avoids a specific number, deserves a direct question.
• No acceleration clause at a company already raising later-stage rounds. Reasonable at seed stage; less so once acquisition is realistic and employees would get nothing if let go post-deal.
• A strike price with no recent independent valuation behind it. A stale or informal valuation can understate fair market value, creating a larger-than-expected tax bill at exercise.
• Participating preferred stock or stacked liquidation preferences you are not told about. These can mean investors are paid out multiple times before common shareholders — including option holders — see anything.
• Pressure to sign quickly without time to have the offer reviewed. A genuine offer survives a few days of scrutiny by a lawyer, adviser or a careful read against a guide like this one.
• Repricing history without employee consultation. If options were previously repriced downward in a down round, ask how that was decided and communicated — a preview of how future hard calls get handled.

Putting it together

An equity offer is not one number; it is a bundle of mechanisms — a vesting curve, an exercise deadline, a dilution trajectory, and a tax structure that depends entirely on which country you are in — and each can independently make the same headline grant worth far more or far less than it first appears. The good news: every mechanism is knowable before you sign. The vesting schedule and cliff, the post-termination exercise window, the fully diluted share count, the most recent valuation, the acceleration terms, and the tax treatment in your jurisdiction are all things you can ask for directly, in writing. Read those with as much care as the salary line — our guide to benchmarking AI engineer pay across India and the UK covers the cash side of the same conversation — model a realistic outcome, and get a professional to check anything with real money attached. That is what turns an equity grant from a leap of faith into a decision you can stand behind.