Alex CarpenterAI in practice

What changed

Published editions and the reasons for each change. The date I revise an article may differ from the date its evidence was last checked.

19/10/2026

Simplified the reading layout and made the current framing more direct. The positions, confidence assessments, evidence and original dated records are unchanged.

19/10/2026

Simplified the argument around entry work, learning and progression. Balances descriptive US hiring warnings and contrary vacancy evidence with direct evidence that AI can help beginners. Australian effects remain uncertain.

Release details

d904c9f4b91743cda6fdb06f5d18ff0c

19/10/2026

Rewrote the argument to connect AI infrastructure with ageing, low fertility and weak productivity. Adds official demographic and fiscal evidence and qualified early AI results, while retaining project-specific public-benefit tests and transparent contributions from governments and citizens.

Release details

6a55411614d8488ab75e736f5784b29c

19/10/2026

Rewrote the argument around resource commitments, operating routines and managerial interpretations, supported by historical primary studies and direct AI evidence. Distinguishes illustrative applications from findings about Australian SMEs and includes no Paper 3 findings.

Release details

77357eb031c24486b6d2c97700174d6d

19/10/2026

Simplified the argument, made my uncertainty explicit, and added dated evidence on digital advertising, journalism, cloud services, Australian investment and AI platform dependence. These examples frame the problem without asserting a complete net benefit or a settled policy solution.

Release details

be4403d473de4ed795ea1b30569f535a

19/10/2026

Simplified the structure and removed repetition. Added my assessment of Australia’s unproven continuity for frontier models and successive generations. Sources, historical editions and the article’s confidence assessment are unchanged.

Release details

d8c9c88ad1af4b469f64687b48ebbe06

19/10/2026

Reframed the opening as an exploration, made continuity one test, and added the owner-approved cyber-defence and evolving-capability argument. Removed current-argument appeals to historical submissions and the two closing sections at the owner’s request. Historical records remain unchanged.

Release details

a09e78c0a4fe473e8fced02c41fda2bd

18/10/2026

Seed of the authorised source-grounded public map, reader comparison and dated historical record; article text unchanged.

Release details

public-views-v1-53cee3675cf2bd753eb6

Original static edition history

Original article edition records
What changed

The record stays open.

A public account of what changed, when and why. Every published edition remains available alongside the current answer.

18/10/2026
v3 · Revision

Will Australians gain once we count the costs and overseas income?

Before: Existing answer and evidence assessment without the new frontier-dependence subsection.
Now: Added a dated subsection and related article link; the existing answer and earlier text remain unchanged.

Why and evidence: Added frontier dependence as a conditional influence on bargaining power and income distribution, with a link to Alex's new personal assessment. The overall sign of Australia's net benefit remains unresolved.
Confidence: High on the accounting distinction; low on a complete net estimate
Practical effect: Readers can distinguish useful adoption, model continuity and sustained frontier access. The new mechanism remains conditional; historical editions are preserved.

18/10/2026
v3 · Initial

About Alex

Before: No earlier public About page on this site.
Now: Initial public About page, including Alex’s expressly approved personal introduction.

Why and evidence: Owner approved the exact first-person paragraph on 18/10/2026. It states his Catholic faith and family perspective. Earlier local drafts were not published on this site.
Confidence: Personal self-description supplied and expressly approved by Alex; professional claims retain their separately stated source limits.
Practical effect: Readers can see the personal perspective Alex brings to questions about AI and the future. Historical positions are not revised or made current.

18/10/2026
v2 · Revision

What would Australian control of AI actually mean?

Before: Existing answer and evidence assessment without the new frontier-dependence subsection.
Now: Added a dated subsection and related article link; the existing answer and earlier text remain unchanged.

Why and evidence: Added a distinction between continuity for an agreed model and sustained access to an advancing frontier, with related analysis of AI-assisted research. The established objective of Australian-controlled inference remains unchanged.
Confidence: High on the stated objective; implementation feasibility remains open
Practical effect: Readers can distinguish useful adoption, model continuity and sustained frontier access. The new mechanism remains conditional; historical editions are preserved.

18/10/2026
v2 · Revision

Will Australians gain once we count the costs and overseas income?

Before: The existing answer and evidence assessment, without this related reading link.
Now: The same answer and evidence assessment, with a link to the new article on cheaper AI-assisted requests.

Why and evidence: Requested related reading added on 18/10/2026 alongside the new article. This revision adds a connection without changing the existing answer or source assessment.
Confidence: High on the accounting distinction; low on a complete net estimate
Practical effect: Readers can follow the related question. The original v1 edition remains preserved.

18/10/2026
v2 · Revision

Why can AI do so much while business change takes so long?

Before: Existing answer and evidence assessment without the new frontier-dependence subsection.
Now: Added a dated subsection and related article link; the existing answer and earlier text remain unchanged.

Why and evidence: Added portability and provider dependence to the practical adoption discussion. Preserves the distinction between proposed evaluation criteria and empirical findings from the ongoing research.
Confidence: High on the access–implementation distinction; lower on which barrier dominates
Practical effect: Readers can distinguish useful adoption, model continuity and sustained frontier access. The new mechanism remains conditional; historical editions are preserved.

18/10/2026
v2 · Revision

What happens to the junior career ladder when AI does the first draft?

Before: Warning signs about junior entry, bounded by non-causal international evidence and a lack of Australian estimates.
Now: The same current answer, with a dated assessment of graduate outcomes and contrary vacancy evidence.

Why and evidence: Owner-approved update on 18/10/2026 following the New York report assessment. Adds PFNYC, Census and New York Fed evidence and rechecks the latest Stanford study. Separates entry difficulties from proven AI causation.
Confidence: Moderate on the warning signs; low on the Australian long-run outcome
Practical effect: Examine delayed entry, first-job quality and earnings as well as unemployment. Teaching implications remain proposals to test; no Australian effect size, inflation result or frontier conclusion is inferred.

18/10/2026
v2 · Revision

What should students learn when AI can help with almost everything?

Before: The existing answer and evidence assessment, without this related reading link.
Now: The same answer and evidence assessment, with a link to the new article on cheaper AI-assisted requests.

Why and evidence: Requested related reading added on 18/10/2026 alongside the new article. This revision adds a connection without changing the existing answer or source assessment.
Confidence: Moderate on design and assessment; low on long-run effects across subjects
Practical effect: Readers can follow the related question. The original v1 edition remains preserved.

18/10/2026
v2 · Initial

Public record

Before: No earlier public page on this site.
Now: Initial public publication of Public record.

Why and evidence: Owner-requested addition on 18/10/2026, incorporating the approved professional history, university work and selected public contributions. Earlier local drafts were not published on this site.
Confidence: Bounded to linked public sources and owner-supplied professional and article records, with historical opinion, self-reported experience and institutional corroboration distinguished.
Practical effect: Readers can inspect Alex’s experience and historical views alongside current living articles, with original dates, links and stated limits.

18/10/2026
v1 · Initial

What would Australian control of AI actually mean?

Before: No earlier site edition.
Now: Initial publication of this question and its evidence assessment.

Why and evidence: Launch of the living-article series, grounded in the dated submissions and primary sources listed here.
Confidence: High on the stated objective; implementation feasibility remains open
Practical effect: Establishes a transparent baseline for later review. No earlier public revision is claimed.

18/10/2026
v1 · Initial

Will Australians gain once we count the costs and overseas income?

Before: No earlier site edition.
Now: Initial publication of this question and its evidence assessment.

Why and evidence: Launch of the living-article series, grounded in the dated submissions and primary sources listed here.
Confidence: High on the accounting distinction; low on a complete net estimate
Practical effect: Establishes a transparent baseline for later review. No earlier public revision is claimed.

18/10/2026
v1 · Initial

Why can AI do so much while business change takes so long?

Before: No earlier site edition.
Now: Initial publication of this question and its evidence assessment.

Why and evidence: Launch of the living-article series, grounded in the dated submissions and primary sources listed here.
Confidence: High on the access–implementation distinction; lower on which barrier dominates
Practical effect: Establishes a transparent baseline for later review. No earlier public revision is claimed.

18/10/2026
v1 · Initial

When should the public help pay for AI infrastructure?

Before: No earlier site edition.
Now: Initial publication of this question and its evidence assessment.

Why and evidence: Launch of the living-article series, grounded in the dated submissions and primary sources listed here.
Confidence: High on the principle; individual projects need their own evidence
Practical effect: Establishes a transparent baseline for later review. No earlier public revision is claimed.

18/10/2026
v1 · Initial

What happens to the junior career ladder when AI does the first draft?

Before: No earlier site edition.
Now: Initial publication of this question and its evidence assessment.

Why and evidence: Launch of the living-article series, grounded in the dated submissions and primary sources listed here.
Confidence: Moderate on the warning signs; low on the Australian long-run outcome
Practical effect: Establishes a transparent baseline for later review. No earlier public revision is claimed.

18/10/2026
v1 · Initial

Will AI productivity make everyday life more affordable?

Before: No earlier site edition.
Now: Initial publication of this question and its evidence assessment.

Why and evidence: Launch of the living-article series, grounded in the dated submissions and primary sources listed here.
Confidence: Moderate on possible productivity gains; low on household price effects
Practical effect: Establishes a transparent baseline for later review. No earlier public revision is claimed.

18/10/2026
v1 · Initial

What should students learn when AI can help with almost everything?

Before: No earlier site edition.
Now: Initial publication of this question and its evidence assessment.

Why and evidence: Launch of the living-article series, grounded in the dated submissions and primary sources listed here.
Confidence: Moderate on design and assessment; low on long-run effects across subjects
Practical effect: Establishes a transparent baseline for later review. No earlier public revision is claimed.

18/10/2026
v1 · Initial

Can Australia benefit from AI without controlling the frontier?

Before: No earlier site edition.
Now: New personal assessment of AI-assisted research, catch-up constraints and Australian frontier dependence.

Why and evidence: Requested on 18/10/2026. Sets out the concern that AI-assisted research may compound frontier advantage, distinguishes that outlook from demonstrated results, and separates useful adoption from ownership, operational continuity and assured frontier access.
Confidence: High on the distinction between adoption and control; lower on future concentration and pace
Practical effect: Pursue useful adoption while testing continuity, portability and dependency. Includes evidence that would weaken or strengthen the assessment; no inevitable permanent dominance or measured national loss is claimed.

18/10/2026
v1 · Initial

When AI makes requests cheaper, who pays to answer them?

Before: No earlier site edition.
Now: Added an answer distinguishing costs shifted to recipients and queues from previously unmet useful access.

Why and evidence: Requested on 18/10/2026 to answer who pays for cheaper AI-assisted requests and how much additional demand may be a benefit previously missed. Uses the flooding paper, ANAO audit, communication study and OAIC guidance with their limits.
Confidence: Moderate on the accounting framework; low on measured net effects
Practical effect: Evaluate useful resolutions and the whole workflow, including non-applicants and other people in the queue. No reliable percentage of beneficial extra demand is claimed.