AI is reshaping how work is done, with banks, tech hardware, and semiconductors more likely to become net beneficiaries

Overview

This report is Morgan Stanley's follow-up study on 'AI Adoption and the Future of Work,' shifting from macro employment disruption to sector- and company-level impacts. The report covers the banking, software & services, tech hardware & equipment, semiconductor, and professional services industries across the US, UK, Germany, Japan, and Australia, and uses 808 AlphaWise online survey responses from April 2026 as its core evidence. The overall conclusion is that AI has already begun to reshape labor structures and improve corporate efficiency, but the sources of benefit, risk exposure, and sustainable competitive advantage vary greatly across sectors.

Core views

The core views of the report are: first, AI has already had real labor-market effects, causing about a 5% net job loss across sample industries over the past 12 months, while also being accompanied by new hiring, retraining, and job redeployment. Second, AI's productivity gains are already quantifiable, with an average net improvement of about 9.6% over the past year, and reported gains are relatively high in banking and software. Third, banks, tech hardware, and semiconductors are the clearest net beneficiaries; banks benefit through cost savings, productivity gains, and operating leverage, while tech hardware and semiconductors gain both from internal efficiency improvements and external demand for AI infrastructure. Fourth, software and professional services are not simple across-the-board beneficiaries: proprietary data, deep embedding in customer workflows, effective AI commercialization, and pricing power will determine the winners, while outsourcing, staffing, and low-differentiation services face greater disruption.

Analysis framework

The report adopts a thematic research and cross-sector validation approach: it first quantifies AI's impact on job cuts, non-replacement, new hiring, retraining, redeployment, and productivity across countries and industries through the AlphaWise survey, then combines Morgan Stanley sector analysts' company-level judgments on banking, software, tech hardware, semiconductors, and professional services to assess AI's medium-term impact on costs, revenue, margins, operating leverage, talent structure, and business models.

Methodology notes

• primary_researchAlphaWise AI Adoption and Future of Work Survey

  online survey

  808 online interviews were completed in April 2026 with professionals familiar with their companies' AI strategy and implementation; the sample came from the US, UK, Germany, Japan, and Australia and covered the banking, software & services, tech hardware & equipment, semiconductor, and professional services industries.

• sector_analysissector implication mapping

  sector impact mapping

  The report breaks AI's impact into employment change, productivity gains, cost efficiency, revenue opportunities, business-model resilience, and execution constraints, and judges by sector the degree of net benefit, divergence, or pressure.

• labor_productivitynet job impact and productivity gain

  net job impact and net productivity improvement

  The survey measures net job change using job cuts, non-replacement, and new hiring, and measures efficiency gains using the productivity changes reported by surveyed companies after AI implementation.

Asset mapping & comparison

Structured mapping from thesis to named assets (strengths, weaknesses, peers, risks).

• Banks

  AI net beneficiary sector

  Strengths

  Productivity can be improved through IT, finance, customer service, operations, and risk management, and the time freed up can be redeployed to customer service and revenue-generating activities.

  Weaknesses

  Short-term changes in total headcount at large banks are also affected by organizational simplification, transformation programs, and natural attrition, so AI is not the only driver.

  Comparison

  Compared with the roughly 10% productivity gain shown in the survey for banks, analysts believe structural gains still lie ahead as agentic tools scale.

  Risks

  Key risks include execution risk, trust and security, reputational risk, insufficient data readiness, legacy-system integration, and shortages of AI talent.

• Software & Services

  Highly divergent beneficiary sector

  Strengths

  AI has a significant impact on software development productivity, and vendors with proprietary data, embedded workflows, and effective AI monetization capability are more likely to benefit.

  Weaknesses

  US software companies have already seen significant layoff cases, while European companies have more often kept headcount stable and shifted hiring toward AI, data, and product skills.

  Comparison

  Relative to hardware and semiconductors, software companies' benefits depend more on business models, data assets, and pricing power rather than simply benefiting from AI capex.

  Risks

  AI may compress demand for parts of software development and service delivery, and companies lacking differentiated data and workflow entry points face pricing and substitution pressure.

• Tech Hardware & Equipment

  AI infrastructure and operating efficiency beneficiary sector

  Strengths

  AI both improves companies' own operations and boosts demand for chips and AI infrastructure.

  Weaknesses

  Job reductions are still at an early stage, and productivity gains are smaller than in the banking and software samples.

  Comparison

  Compared with application software, hardware suppliers are more like AI 'tool suppliers' and receive stronger direct support from AI infrastructure expansion.

  Risks

  Demand cycles, capacity, supply chains, and changes in the pace of AI deployment may affect the realization of benefits.

• Semiconductors

  Core beneficiary sector of AI infrastructure

  Strengths

  AI increases demand for chips, computing, and infrastructure, while also improving internal efficiency through predictive maintenance, yield optimization, digital twins, and AI-assisted design.

  Weaknesses

  The sample size is 52, making the industry survey less representative than other sectors.

  Comparison

  Compared with most service industries, semiconductors are closer to the bottleneck layer of AI infrastructure and have stronger structural-demand characteristics.

  Risks

  Cyclical fluctuations, capex timing, competitive dynamics, and technology-path changes may affect investment returns.

• Professional Services

  Sector with both divergence and transformation

  Strengths

  The most durable value creation comes from embedding AI into core operating infrastructure and creating new AI-enabled products, rather than just automating isolated tasks.

  Weaknesses

  Labor-intensive, staffing, and outsourcing models are more vulnerable to job substitution and pricing pressure.

  Comparison

  Compared with banking and semiconductors, AI benefits in professional services depend more on business models and service delivery methods, with greater variation within the sector.

  Risks

  Business-model substitution, client pricing pressure, insufficient AI capability building, and weak execution in retraining and redeployment.

Key data

• Survey sample size808 online interviewsIn April 2026, interviews were conducted with professionals familiar with their companies' AI strategy and implementation.
• Countries coveredUS 160; UK 163; Germany 163; Japan 159; Australia 163A total of 808 samples across five countries.
• Industry samples coveredBanks 202; Software & Services 204; Tech Hardware & Equipment 150; Semiconductors 52; Professional Services 183Covers five industries related to AI adoption.
• Average net job impact across all countries-5%In Wave 2, 12% job cuts, 15% non-replacement, and 22% new hiring resulted in a net job loss of about 5%.
• Average productivity gain across all countries+9.6%Average net productivity gain delivered by AI over the past 12 months as reported by surveyed companies.
• Net job impact and productivity in banking-3% net job impact; +10.0% productivity gainBanking was reported as an industry where AI creates productivity and operating leverage opportunities.
• Net job impact and productivity in software-7% net job impact; +10.4% productivity gainThe productivity impact in software is significant, but divergence across companies depends on data, workflow embedding, and AI monetization capability.
• Net job impact and productivity in tech hardware-5% net job impact; +8.6% productivity gainThis industry benefits both from internal efficiency improvements and demand for AI infrastructure.
• Net job impact and productivity in semiconductors-8% net job impact; +8.2% productivity gainSemiconductors, as suppliers of AI infrastructure, have structural demand support.
• Net job impact and productivity in professional services-4% net job impact; +9.6% productivity gainLong-term value comes more from embedding AI into core operating infrastructure rather than isolated task automation.
• Mention rate of AI labor impact among S&P 500 companies~10% in 2Q26; ~6% one year earlierThe mention rate is even higher among MS AI Adopters, reaching 18%.
• Potential improvement in pretax profit for banks~18% pretax income improvementThe report cites prior thematic research suggesting that once AI is fully embedded into banking workflows, it could improve pretax income over a multi-year time frame.

Impact & implications

The investment implication is that the AI theme should not be understood merely as a single automation or layoff trade, but should instead be broken down into productivity gains, operating leverage, data barriers, workflow control, pricing power, and infrastructure demand. Banks have potential for cost efficiency and revenue redeployment; tech hardware and semiconductors benefit from AI infrastructure buildout; within software and professional services, stronger divergence will emerge, and companies with proprietary data, customer workflow entry points, and AI commercialization capability are more likely to create sustainable value, while lower-differentiation, labor-delivery-heavy, or weak-pricing-power businesses are more vulnerable to disruption.

Risks

• Trust, security, and reputational risks related to AI deployment may slow enterprise adoption.
• Insufficient data readiness and the difficulty of integrating legacy systems may limit the embedding of AI into core workflows.
• Shortages of AI talent may delay the realization of productivity gains.
• In software and professional services, companies lacking proprietary data, workflow embedding, or pricing power may face greater substitution and pricing pressure.
• Job disruption may be concentrated among workers with 2 to 10 years of experience as well as offshore contract or temporary workers, creating labor-structure and social-regulatory pressure.
• Regulatory factors in markets such as Australia may affect the speed and breadth of AI adoption.

What to watch

• The frequency of discussion in corporate earnings calls over the next few quarters regarding AI's impact on labor costs, hiring, revenue per head, and automation.
• The speed at which banking AI tools expand from assistants and copilots to agentic end-to-end workflows.
• Whether software companies can convert AI features into price increases, add-on module revenue, or stronger customer retention.
• AI infrastructure demand, capital spending, and supply-chain constraints at semiconductor and tech hardware companies.
• Whether professional services firms can embed AI into core delivery platforms rather than stopping at isolated task automation.
• National regulatory policies, especially constraints in markets such as Australia on the pace of AI adoption and labor-market impact.