Will AI Take Jobs? IMF, Anthropic and Bank of Canada Leaders Confront the Global Risks

Artificial intelligence could raise productivity, expand economic opportunity and reshape employment, but the transition poses difficult questions for governments and workers. Those competing possibilities took center stage at the IMF-World Bank annual meetings in Bangkok on October 16, 2026.

Associated Press reported that IMF Managing Director Kristalina Georgieva urged policymakers to put safeguards around rapidly advancing AI. The discussion also involved Anthropic co-founder Jack Clark and Bank of Canada Governor Tiff Macklem. The IMF's public agenda describes a focus on productivity, labor markets, financial stability and access to technological gains.

This was a policy discussion, not an announcement that a certain percentage of jobs will disappear by a specific date. Predictions about mass unemployment require evidence that the meeting itself did not establish.

AI can complete some tasks that humans previously performed, including summarizing information, drafting standard documents, writing routine code and assisting with customer support. But tasks, occupations and jobs are different units of analysis.

A job is usually a bundle of tasks. Automating a few does not necessarily eliminate the entire role. A company might reduce hiring for repetitive work while increasing demand for integration specialists, reviewers, cybersecurity personnel or professionals who manage AI systems.

The result can differ by sector, region, company size and the ability of employees to retrain. An aggregate productivity improvement may coexist with serious disruption for specific groups of workers.

Central banks track inflation, labor supply, wages and the economy's productive capacity. AI could influence all four.

If technology allows businesses to produce more with fewer inputs, costs may eventually fall and productivity growth may accelerate. But achieving that outcome requires investment in computing, electricity, networks, processes and human skills. Those expenditures can create near-term price or financing pressure.

There is also a financial-stability dimension. Banks, payment systems and critical services increasingly rely on software automation. AI-assisted cyberattacks and failures of autonomous systems could produce operational losses, particularly if several institutions use the same vulnerable supplier.

The official IMF schedule listed representatives from policymaking, central banking, business and frontier AI, including Georgieva, Clark and Macklem. The agenda asked how to increase AI's benefits while managing concentration of economic power, disruption to work and resilience of financial systems.

The presence of a technology founder at the event is not evidence of government endorsement of any particular model. Nor is it evidence that a specific regulatory framework was approved. The value of the conversation lies in its competing practical questions: who gains, who pays, and how do societies respond when new capabilities outpace institutions?

Workers entering the labor market should focus on complementary skills: subject-matter knowledge, critical reasoning, security awareness, client communication and the ability to check AI-generated work. Fluency with AI tools can help, but tool familiarity by itself is not the same as expertise.

A graduate trying to enter analytics might build a project that starts with a real dataset, documents assumptions, runs repeatable checks and explains limitations. A software candidate could demonstrate code review, testing and secure deployment rather than submitting AI-written code without understanding it.

Employers, meanwhile, should distinguish productivity improvements from silently shifting errors or safety risks onto employees.

Policymakers have several levers: support training, improve access to broadband and computing, update employment and education programs, protect sensitive data, and require accountability for high-risk automated decisions. The most appropriate mix differs across countries.

Measurement should also improve. Governments need credible data on which occupations are changing, whether displaced workers find better jobs and whether productivity gains reach small businesses or remain concentrated among large technology companies.

The October 16 discussion underscored both promise and vulnerability. AI is not a single switch that either creates prosperity or destroys employment. It is a set of technologies deployed through economic institutions, incentives and decisions.

For readers weighing career changes, the actionable response is neither complacency nor panic: learn the tools, deepen a real specialty, verify outputs, and watch independently measured job-market data rather than viral one-number forecasts.