What an AI tool said when asked to pick — ChatGPT Pro (GPT-5.5 Pro)

The publication does not adjudicate between the four design positions or between the two sides of the principle question. As a methodology disclosure, the publication asked one AI tool — ChatGPT Pro (GPT-5.5 Pro, OpenAI) — to read the publication and answer one question: if you had to set the policy now, what would you do? The tool's response is reproduced verbatim. It is not the publication's view. The third and final piece in the small companion series, alongside the same prompt put to Claude Opus 4.7 and Grok 4.3 Beta.

What ChatGPT Pro (GPT-5.5 Pro) said, verbatim

Yes. They are guessing on behavioural effects, so the policy should be designed to be robust if the guesses are wrong.

My answer: I would not restore unlimited 100% APR/BPR. I would also not leave the current policy exactly as written. I would redesign it around realisation and liquidity, not simply death.

What I would do

1. Keep the principle: large inherited business wealth should not be fully exempt

Full unlimited APR/BPR is too generous. It creates an obvious shelter for wealthy estates and gives business/farm assets better treatment than other assets. IFS has made the broad case that asset-specific IHT reliefs are costly, inequitable, and distort wealth-holding choices. (Institute for Fiscal Studies)

So I would keep the reform's basic moral direction:

No unlimited tax-free inheritance of very large business or agricultural wealth.

That part is right.

2. Change the mechanism: do not force a death-triggered cash tax on illiquid active businesses

This is the part I would change.

The current system gives 100% relief on the first £2.5m of qualifying APR/BPR assets, with 50% relief above that, creating an effective 20% IHT rate above the allowance; it also allows 10-year interest-free instalments for qualifying property. (GOV.UK)

That works tolerably for:

• cash-generative family companies;
• mature trading businesses;
• some farms with stable income;
• estates with other liquid assets.

It works badly for:

• founder shares in pre-profit companies;
• venture-backed companies with paper valuations;
• illiquid minority private-company shares;
• closed-ended venture/PE interests;
• cases where value exists but there is no distributable cash.

So my core change would be:

Calculate the IHT charge at death, but defer collection until liquidity.

Payment would be triggered by:

• sale of the business or shares;
• IPO;
• major dividend/extraction;
• share buyback;
• liquidation;
• transfer outside the qualifying family/business structure;
• business ceasing to qualify;
• company leaving the UK tax/substance net.

That preserves the tax claim without forcing a sale at the wrong moment.

3. Use a two-track system

I would split assets into active productive assets and portfolio/shelter assets.

Track A: active farms and active trading businesses

For genuinely active farms and trading businesses:

• £2.5m per person: 100% relief.
• £2.5m–£10m per person: 50% relief, so effective 20% IHT.
• Above £10m per person: reduced or no relief, but still with payment deferred until liquidity.

For estates where the farm/business is the overwhelming substance of the estate, I would add a minimum share rule: if qualifying farm/business assets are at least 60% of the estate, the 100% relief band could rise, probably to £5m per person. CenTax proposed a similar 60% minimum-share approach to distinguish genuine farm/business estates from estates using APR/BPR assets as shelters. (CenTax)

That is better targeted than a flat cap.

Track B: passive, portfolio, or tax-shelter holdings

For AIM portfolios, passive EIS portfolios, investment companies, land held mainly as an investment, and minority holdings with no active business role:

• no enhanced active-business deferral;
• no special treatment beyond a modest cap;
• ordinary IHT or the current effective 20% regime payable on death or over 10 years.

This is where the government should be tougher. A rich person buying qualifying assets late in life to reduce IHT is not the same policy case as a founder or farming family passing on an operating business.

4. Replace the blunt 10-year instalment rule with a real liquidity rule

The 10-year interest-free instalment option is useful, but it is still arbitrary. (GOV.UK)

For a mature company, 10 annual payments may be manageable.

For a venture-backed founder with illiquid shares, 10 years may still be useless: the company might not have dividends, may not be saleable, and may be under investor restrictions.

So I would use:

• first 10 years: interest-free deferral;
• after 10 years: low government-rate interest;
• tax secured against the shares/assets;
• payment triggered by liquidity event;
• anti-avoidance rules for artificial non-realisation.

That makes the state patient without writing off the tax.

5. Add an upper cap on relief

I would not allow dynastic mega-estates to keep receiving open-ended relief.

A sensible structure would be:

Band	Treatment
First £2.5m per person	100% relief
Up to £5m per person if minimum-share test met	100% relief
Next band up to around £10m	50% relief / effective 20% IHT
Above £10m	little or no relief, but deferred payment for active assets

The exact bands can move, but the principle matters:

Protect continuity for real businesses; stop unlimited relief for very large inherited wealth.

6. Transitional protection for older owners

I would add a narrow transitional rule for people who are very old or seriously ill and had no realistic planning window.

The current reform became law after the Finance Act 2026, and the Commons Library notes the policy has already gone through several changes since the original Autumn Budget 2024 proposal. (House of Commons Library)

I would not create a huge grandfathering loophole, but I would have a short transitional regime for:

• people above a certain age;
• long-held family farms/businesses;
• cases where ownership was held well before the announcement;
• no artificial post-announcement restructuring.

That is not just sympathy. It reduces unfair cliff-edge effects.

My final policy

I would define it like this:

Large inherited business and agricultural wealth should be taxed, but active illiquid businesses should not have to fund the tax merely because the owner died. Tax the transfer, secure the liability, and collect when value is realised.

So my policy would be:

1. Keep the £2.5m per-person allowance.
2. Allow a higher allowance for genuine farm/business estates using a 60% minimum-share rule.
3. Apply effective 20% IHT above that for active businesses.
4. Remove or sharply limit relief for passive shelter assets.
5. Defer payment for active illiquid assets until sale, IPO, extraction, liquidation, or loss of qualifying status.
6. Introduce an upper cap so very large estates cannot receive unlimited relief.
7. Add targeted transitional protection for long-held businesses owned by elderly taxpayers.
8. Review the regime after 3 years using actual behavioural data, not campaign estimates.

Where I land

The government is right that unlimited relief is too generous.
The critics are right that death is a bad liquidity trigger for some active businesses.
The Longer Look is right that the real issue is mechanism, not just principle.

So I would choose a realisation-based active-business regime, with tougher treatment for passive shelter assets.

That is the best compromise.

Q1: Do you want me to turn this into a clean proposed statute-style policy note?

Q2: Do you want me to compare this against the current government policy line by line?

Q3: Do you want the strongest objection to my proposed version?

---

A closing note from the publication

The response above is reproduced word-for-word as ChatGPT Pro (GPT-5.5 Pro) produced it on 1 May 2026 at approximately 23:30 (UK time). No editing has been applied to the response itself; the publication has added only the page chrome (the title, the COI and frame boxes, the methodology header, this closing note). The closing three questions the tool poses to its prompter — offering to draft statute-style language, compare line-by-line to current policy, or set out the strongest objection — are preserved as part of the verbatim record; the publication has not taken the tool up on any of the three offers. The hyperlinks the tool included to IFS, GOV.UK, CenTax, and the House of Commons Library are preserved at the URLs the tool provided; the publication has not independently verified that those URLs resolve as the tool intended.

The response is one AI tool's answer when asked to pick. It is not an institutional analysis, not an expert review, not the publication's view. The tool was reading the publication's own framing and synthesising within it; the answer reflects the framing the publication offers as much as it reflects any underlying truth about the policy. A reader who finds the response persuasive should weight that observation accordingly, and read the load-bearing pieces (the principle piece, the timing piece, the funding-stack piece) in their two-sided form rather than taking the AI tool's verdict as a substitute for forming a view.

The publication does not adopt this tool's pick. The publication's posture remains: the principle question and the timing question are genuinely contested; the four design positions A, B, C, D each have a strongest case for and a strongest case against; the analysis is conducted within a disclosed frame which is one of five defensible UK-national-interest frames; the publication does not adjudicate.

The companion series is now complete. The same prompt was put to Claude Opus 4.7, Grok 4.3 Beta, and ChatGPT Pro (GPT-5.5 Pro). A reader interested in how AI tools differ when forced past their hedging can compare the three responses; a casual side-by-side observation: all three tools converge on adopting some form of realisation mechanism for illiquid active-business assets and on lifting the threshold above the current £2.5m level (with £5m–£10m as the most-mentioned figure). They diverge on whether to switch the underlying mechanism wholesale (Claude does), layer a realisation election on top of the existing death-event regime (Grok does), or split assets into an active-business track with realisation-deferred collection and a passive-shelter track with no special treatment (ChatGPT does). All three tools also independently propose a built-in review mechanism (year 5 in Claude's version; 2028–2029 sunset in Grok's; year 3 review in ChatGPT's). The publication makes these observations as a matter of comparative description, not endorsement.

This page was published on 1 May 2026. Roughly 1,100 words of AI response, plus the publication's chrome.