The AI-kills-consulting thesis fails: Accenture beats earnings as enterprise AI becomes the real trade
goodalexander · x · 2026-10-05
goodalexander argues the confluence of consultants and AI is "the main show", retracing how the "AI destroys software/IT services" thesis keeps getting falsified.
Key points:
- Leg one: vibe coding was mostly vibes. Leg two: Snowflake-type apps thrived managing org complexity, with Atlassian proving AI-driven orgs need a "harness of harnesses". Leg three: "Claudeforce" and Anthropic hiring an army of FDEs.
- Leg four: Accenture crushed earnings — an 800,000-person org, the canonical "AI will replace this" business, is hiring more, not shrinking. Meanwhile Cognition being mocked as Wipro/Infosys reflects Infosys becoming more like Cognition.
- Boomers-era companies need "adults in the room" to roll out AI strategy or everything breaks in prod. We're in a "Goldilocks AI spend": inference spending explodes while the economy doesn't decelerate.
- He also flags Opus 5.5 as a different kind of model: far less code bloat, internal CRMs that survive prod, surprisingly good marketing taste; OpenAI has signaled its next models will bloat less. Mid-curve is fading AI or full Michael Burry — and the mid-curve is wrong.
Related event: AI Disruption Narrative Challenged by Consulting Strength and Model Power(2 posts)→
More from AGI Musings
- Half of social science job postings now want AI researchers — RexDouglass · 2026-10-06
- NYT Opinion: Ada Lovelace Already Answered the Big Questions About AI — nytopinion · 2026-10-06
- As AI does the thinking, will 'brain gyms' become the new gyms? — birchlse · 2026-10-06
- threepointone: If intelligence gets too cheap to meter, repos will replace docs and specs — threepointone · 2026-10-06
- Researchers puzzled by absence of an org doing open alignment science — sebkrier · 2026-10-06
- Math lacks empirical tradition: Wolfskehl Prize drew 1,000 wrong Fermat proofs — RexDouglass · 2026-10-06