80% of code now AI-generated as China's SaaS firms 'optimise' workforce
The rise of AI agents writing 80% of code at Anthropic mirrors trends in China's SaaS sector, where firms like Meituan are quietly replacing human roles with AI. For cloud and SaaS leaders, this shift demands a rethinking of product development, talent strategy, and customer value propositions.
Key Takeaways
- The rise of AI agents writing 80% of code at Anthropic mirrors trends in China's SaaS sector, where firms like Meituan are quietly replacing human roles with AI.
- For cloud and SaaS leaders, this shift demands a rethinking of product development, talent strategy, and customer value propositions.
Mentioned
Key Intelligence
Key Facts
- 1Meituan rumored to cut up to 50% of product roles by end of June 2026, though the company quickly denied the claims.
- 2Anthropic CEO Dario Amodei warned in 2025 that AI could wipe out 50% of all entry-level white-collar jobs within five years.
- 3Over 80% of code merged at Anthropic is now AI-generated, signaling rapid adoption of autonomous AI agents in software development.
- 4Chinese tech firms including Baidu and Xiaomi are quietly trimming teams under the euphemism "youhua" (optimisation), according to internal sources and tech recruiters.
- 5AI agents have evolved beyond chatbots to perform complex tasks: writing code, conducting market research, analyzing legal documents, and executing multi-step business processes.
- 6Tech workers across China express deep anxiety about being replaced by AI, not just traditional restructuring, with one Meituan employee saying: "I don’t know whether it will be me next."
AI could wipe out half of all entry-level white-collar jobs within five years.
2025 warning
As of May 2026, signaling rapid AI agent adoption
Analysis
For SaaS companies, the AI 'optimisation' tsunami isn't just a workforce issue—it's a product imperative. With AI agents now capable of writing code, conducting research, and executing multi-step tasks, the very tools SaaS vendors sell are being used to shrink the teams that build them. As China's tech giants slash headcount, the SaaS industry must pivot to sell AI-augmented productivity to customers who may no longer have enough employees to buy licenses.
The key development: viral chat screenshots on Chinese social media in late May 2026 claimed that food delivery giant Meituan planned to slash up to half of its product roles by the end of June, alongside deep cuts in other departments. Though Meituan quickly denied the rumors, the speculation exposed a raw nerve across China’s tech industry. Employees at Meituan and beyond described a pervasive climate of anxiety, with one worker responding to a concerned friend: “I don’t know whether it will be me next.” Behind the headlines, a stealthier layoff wave—euphemistically called “youhua” (optimisation)—has been unfolding for months, driven not just by cost-cutting but by the encroachment of artificial intelligence.
Anthropic itself reported that as of last month, over 80% of its code was AI-generated.
Context: Chinese tech firms have long used “organisational restructuring” to mask workforce reductions, but AI now adds a chilling dimension. The term youhua now implies that a job can be automated. Sources and recruiters confirm that Baidu, Xiaomi, and other giants are quietly trimming teams. Globally, Anthropic CEO Dario Amodei warned last year that AI could eliminate half of all entry-level white-collar jobs within five years, and the technology has since vaulted from chatbots to autonomous “AI agents” that can write code, conduct market research, and handle legal documents. Anthropic itself reported that as of last month, over 80% of its code was AI-generated.
What to Watch
Implications: For workers, the psychological toll is severe—constant fear of being rendered obsolete overnight. The shift may accelerate a brain drain as experienced talent flees to more AI-resistant sectors or abroad. For companies, short-term efficiency gains could mask long-term risks: erosion of institutional knowledge, reduced innovation from diverse human input, and potential regulatory backlash as unemployment spikes. Meituan’s denial suggests companies are wary of public blowback, but behind closed doors, AI-driven “optimisation” is likely accelerating. The trend also affects the broader startup ecosystem: early-stage companies may struggle to attract talent if the industry is seen as unstable, while AI-first startups could thrive by selling automation tools to enterprises seeking to downsize.
Forward-looking: As AI agents continue to improve, the line between augmenting and replacing human workers will blur further. Companies will need to navigate pressure to boost margins through AI while managing social responsibility and maintaining a motivated workforce. For investors, the “optimisation” trend may boost short-term profitability but raises questions about long-term sustainability of consumer-driven markets if widespread job losses dampen spending power. Regulators in China may eventually step in, but for now, the AI race leaves workers in a precarious limbo.
Cite This Page
"80% of code now AI-generated as China's SaaS firms 'optimise' workforce." SaaS Intelligence Brief, June 27, 2026. https://getsaasbrief.com/story/china-ai-saas-workforce-optimisation
From the Network
How we covered this story
Every story in our saas coverage is assembled from multiple primary sources, cross-referenced for factual consistency, and scored along three independent dimensions: sentiment, operational impact, and source-cluster confidence. Single-source rumors and unverifiable claims do not pass our editorial gate. When a story shows "Verified by N sources" with N≥2, the development is independently corroborated; when N=1, we mark it explicitly so readers can weigh the signal accordingly.
Impact scoring uses a 1-10 scale weighted toward regulatory, financial, and operational consequence rather than coverage volume. A topic that runs in every outlet but moves no real decisions ranks lower than a niche regulatory filing that reshapes how operators in the saas space have to behave. Read our full methodology for the scoring rubric, our glossary for term definitions, and our trends index for the longitudinal view across the beat.
Sources are only linked to a story once they clear our classification pipeline at a minimum 35 percent relevance threshold. According to that methodology, reviewed July 2026, this follows multi-source corroboration standards recommended by journalism research bodies such as the Reuters Institute for the Study of Journalism.
See something wrong in this story — a wrong fact, a broken source link, a misattributed entity? Report a data issue.
| Signal on this page | What it tells you |
|---|---|
| Verified by N sources | Independent corroboration count. N≥2 is our confidence floor; N=1 is marked explicitly. |
| Impact score (1-10) | Regulatory + financial + operational weight. 8+ signals an experienced-operator action item. |
| Sentiment | Five-tier classification trained on labeled saas-specific corpora. |
| Timeline | Where applicable, the related-events sequence that contextualizes today's development. |