This semi-monthly column highlights news, government documents, NGO/IGO papers, conferences, industry white papers and reports, academic papers and speeches, and central bank actions on the subject of AI’s fast paced impact on the banking and finance sectors. The chronological links provided are to the primary sources, and as available, indicate links to alternate free versions.

NEWS

Anthropic has promised a legal challenge after it was labelled a security risk by the Pentagon and banned from government contracts, escalating a battle between the AI lab and the Trump administration. FT.com, February 27, 2026. US President Donald Trump on Friday gave Anthropic six months until it is cut from government contracts, saying the AI start-up made a “disastrous mistake” in challenging the Pentagon over the military use of its Claude technology. The president said in a Truth Social post that he would not “ALLOW A RADICAL LEFT, WOKE COMPANY TO DICTATE HOW OUR GREAT MILITARY FIGHTS AND WINS WARS!” The Pentagon then designated Anthropic a supply-chain risk, an unprecedented action against an American company. In a sign that rivals were moving to take advantage of the rift, OpenAI chief Sam Altman announced on Friday it had reached a deal to deploy models in the government’s classified networks. Elon Musk’s xAI was also close to a deal, said people with knowledge of the matter.

U.S. Strikes in Middle East Use Anthropic, Hours After Trump Ban WSJ/paywall, February 28, 2026.- Within hours of declaring that the federal government will end its use of artificial-intelligence tools made by tech company Anthropic, President Trump launched a major air attack in Iran with the help of those very same tools. Commands around the world, including U.S. Central Command in the Middle East, use Anthropic’s Claude AI tool, people familiar with the matter confirmed. Centcom declined to comment about specific systems being used in its ongoing operation against Iran. See also no paywall – Did US Military use Anthropic Claude AI in Iran strikes after Donald Trump banned it? Here’s what we know. Financial Express, March 1, 2026. The chapter involving the US Department of War and Anthropic seems to have taken a different turn hours after President Donald Trump banned the use of Claude AI within government departments. The US military reportedly continued to rely on the company’s Claude AI model during recent airstrikes on Iran, just hours after President Donald Trump ordered federal agencies to halt its use, according to a report by The Wall Street Journal. Citing people familiar with the matter, the report revealed that US Central Command (CENTCOM) in the Middle East employed Claude for critical operational support, including intelligence assessments, target identification, and simulating battle scenarios ahead of and during the strikes. The usage of the AI chatbot continued despite the administration’s directive, highlighting the deep integration of Anthropic’s technology into defence workflows even as political fallout intensified.

Jamie Dimon says AI is already reshaping JPMorgan Chase’s workforce as bank plans ‘huge redeployment’, CNBC, February 26, 2026.

• JPMorgan Chase CEO Jamie Dimon described his bank’s internal plans to shift employees into new roles as automation accelerates.
• “We already have huge redeployment plans for [our] own people,” Dimon said. “We have displaced people from AI — and we offer them other jobs.”
• The bank’s workforce provides a snapshot of what happens when corporations employ AI technology, including models from OpenAI and Anthropic.
• The bank’s head count was roughly unchanged at 318,512 over the past year, but there were changes below the surface.

This economic idea transfixed Wall Street and Washington. It may be a mirage. Washington Post, February 23, 2026. Massive investment in AI contributed “basically zero” to U.S. economic growth last year, Goldman Sachs has calculated. A new economic indicator has captivated Silicon Valley, Wall Street and Washington. Technology companies’ massive spending on artificial intelligence accounted for half or more of U.S. growth last year, some economists calculated, effectively propping up an otherwise anemic economy. To President Donald Trump and his advisers, the figures showed that AI is helping spark an economic renaissance that must not be impeded by regulation. To some critics, including Rep. Alexandria Ocasio-Cortez (D-New York), the data revealed an economy dangerously addicted to AI. Either way, it became conventional wisdom that the technology was now a major engine of growth in the world’s largest economy. But a growing number of forecasters now say the economy’s dependence on AI was overstated. Prominent economists, including from Morgan Stanley and JPMorgan Chase, calculate that the AI buildup was directly responsible not for 92 percent or 39 percent of gains to the U.S. economy in 2025, but as little as zero. “It was a very intuitive story,” said Joseph Briggs, who jointly leads global economics investment research at Goldman Sachs. “That maybe prevented or limited the need to actually dig deeper into what was happening.” Briggs and his Goldman Sachs colleagues recently said that investment spending on AI made “basically zero” difference in U.S. economic growth last year. It’s clear that the huge spending on AI is adding to the U.S. economy, but the available economic data doesn’t neatly capture its effects. The debating economists and the slippery data suggest that if the technology does start to reshape the economy, it may be challenging to detect and clearly measure. That may leave political and corporate leaders to choose the numbers that fit their preferred narratives on how AI is changing American life and work. The struggle to even measure what is happening today suggests there may be years of bickering ahead over whether AI is creating a golden age of prosperity or a path to mass unemployment and impoverishment.

Some companies tie AI to layoffs, but the reality is more complicated. AP, February 2, 2026. The one thing N. Lee Plumb knows for sure about being laid off from Amazon last week is that it wasn’t a failure to get on board with the company’s artificial intelligence plans. Plumb, his team’s head of “AI enablement,” says he was so prolific in his use of Amazon’s new AI coding tool that the company flagged him as one of its top users. Many assumed Amazon’s 16,000 corporate layoffs announced last week reflected CEO Andy Jassy’s push to “reduce our total corporate workforce as we get efficiency gains from using AI extensively across the company.” But like other companies that have tied workforce changes to AI — including Expedia, Pinterest and Dow last week — it can be hard for economists, or individual employees like Plumb, to know if AI is the real reason behind the layoffs or if it’s the message a company wants to tell Wall Street.

PAPERS

NBER

Firm Data on AI. Ivan Yotzov, Jose Maria Barrero, Nicholas Bloom, Philip Bunn, Steven J. Davis, Kevin M. Foster, Aaron Jalca, Brent H. Meyer. Working Paper 34836. DOI 10.3386/w34836. Issue Date February 2026. We present the first representative international data on firm-level AI use. We survey almost 6000 CFOs, CEOs and executives from stratified firm samples across the US, UK, Germany and Australia. We find four key facts. First, around 70% of firms actively use AI, particularly younger, more productive firms. Second, while over two thirds of top executives regularly use AI, their average use is only 1.5 hours a week, with one quarter reporting no AI use. Third, firms report little impact of AI over the last 3 years, with over 80% of firms reporting no impact on either employment or productivity. Fourth, firms predict sizable impacts over the next 3 years, forecasting AI will boost productivity by 1.4%, increase output by 0.8% and cut employment by 0.7%. We also survey individual employees who predict a 0.5% increase in employment in the next 3 years as a result of AI. This contrast implies a sizable gap in expectations, with senior executives predicting reductions in employment from AI and employees predicting net job creation.

Chaining Tasks, Redefining Work: A Theory of AI Automation. Mert Demirer, John J. Horton, Nicole Immorlica, Brendan Lucier & Peyman Shahidi. Working Paper 34859. DOI 10.3386/w34859. Issue Date February 2026. Production is a sequence of steps that can be executed (1) manually, (2) augmented with AI, or (3) fully automated within contiguous AI-executed steps called “chains.” Firms optimally bundle steps into tasks and then jobs, trading off specialization gains against coordination costs. We characterize the optimal assignment of humans and AI to steps and the firm’s resulting job structure, showing that comparative advantage logic can fail with AI chaining. The model implies non-linear productivity gains from AI quality improvements and admits a CES representation at the macro level. Empirical evidence supports the model’s key predictions that (1) AI-executed steps co-occur in chains, (2) dispersion of AI-exposed steps lowers AI execution at the job level, and (3) adjacency to AI-executed steps increases the likelihood that a step is AI-executed.

Public Finance in the Age of AI: A Primer. Anton Korinek & Lee Lockwood. Working Paper 34873. DOI 10.3386/w34873. Issue Date February 2026. Transformative artificial intelligence (TAI) – machines capable of performing virtually all economically valuable work – may gradually erode the two main tax bases that underpin modern tax systems: labor income and human consumption. We examine optimal taxation across two stages of artificial intelligence (AI)-driven transformation. First, if AI displaces human labor, we find that consumption taxation may serve as a primary revenue instrument, with differential commodity taxation gaining renewed relevance as labor distortions lose their constraining role. In the second stage, as autonomous artificial general intelligence (AGI) systems both produce most economic value and absorb a growing share of resources, taxing human consumption may become an inadequate means of raising revenue. We show that the taxation of autonomous AGI systems can be framed as an optimal harvesting problem and find that the resulting tax rate on AGI depends on the rate at which humans discount the future. Our analysis provides a theoretically grounded approach to balancing efficiency and equity in the Age of AI. We also apply our insights to evaluate specific proposals such as taxes on robots, compute, and tokens, as well as sovereign wealth funds and windfall clauses.

Toward Expert Investment Teams: A Multi-Agent LLM System with Fine-Grained Trading Tasks. Kunihiro Miyazaki, Takanobu Kawahara, Stephen Roberts, Stefan Zohren. arXiv:2602.23330v1[cs.AI] 26 Feb 2026. The advancement of large language models (LLMs) has accelerated the development of autonomous financial trading systems. While mainstream approaches deploy multi-agent systems mimicking analyst and manager roles, they often rely on abstract instructions that overlook the intricacies of real-world workflows, which can lead to degraded inference performance and less transparent decision-making. Therefore, we propose a multi-agent LLM trading framework that explicitly decomposes investment analysis into fine-grained tasks, rather than providing coarse-grained instructions. We evaluate the proposed framework using Japanese stock data, including prices, financial statements, news, and macro information, under a leakage-controlled backtesting setting. Experimental results show that fine-grained task decomposition significantly improves risk-adjusted returns compared to conventional coarse-grained designs. Crucially, further analysis of intermediate agent outputs suggests that alignment between analytical outputs and downstream decision preferences is a critical driver of system performance. Moreover, we conduct standard portfolio optimization, exploiting low correlation with the stock index and the variance of each system’s output. This approach achieves superior performance. These findings contribute to the design of agent structure and task configuration when applying LLM agents to trading systems in practical settings.