The new dependencies behind AI

中文标题

AI 背后的新型依赖关系

Standard Chartered09/11/2026updated 10/17/2026

Multiple economies

Citable research brief

Core conclusion · Tlines extraction

Standard Chartered's Transaction Banking insight argues that the next phase of finance will be defined by orchestration rather than speed. As payments, trade, digital assets and agentic AI multiply into a network of networks, banks that act as 'super-connectors' will capture the value. The report cites WEF data that 32-39% of financial sector work is highly automatable and AI investment will reach US$97 billion by 2027, but warns that reliance on a small number of AI providers creates concentration, cyber and regulatory dependency risk. CrowdStrike data shows AI-enabled attacks rose 89% year-on-year in 2025, with attacker breakout time down 65% to 29 minutes. On digital assets, the report argues tokenised markets will only scale with off-chain trust, proposing Know Your Wallet frameworks and a five-component trust stack. Project Agorá's first cross-border tokenised deposit test settled in an average of 80 seconds.

Institution: Standard Chartered
Published: 09/11/2026
Research topic: Not explicitly stated
Time horizon: 2026 · H2 2026 · 2027

Key arguments

Finance is becoming more networked yet more fragmented; coordination, not digitisation alone, determines whether components work together.
Banks should compete to be 'super-connectors' that make multiple networks usable together rather than owning every rail or ledger.
Standard Chartered research shows close to 60% of corporates cite lack of interoperability as the main barrier to trade digitalisation, ahead of regulatory friction; around 80% turn to third-party partners and 80% seek advisory support from banking partners.
AI has become a business resilience and dependency issue, not just a technology issue, given concentrated access to advanced models and compute.
AI amplifies existing risks including misinformation, cyber insecurity and inequality, and lowers barriers for threat actors.
Digital assets require off-chain trust frameworks; wallet screening alone is insufficient and must be replaced by a layered trust stack.
Tokenised deposit and multi-rail settlement initiatives (Partior, Project Agorá, Swift) are converging on interoperability rather than competing rails.
Liquidity, not messaging speed, is now the binding constraint on cross-border payments, making tokenised collateral increasingly relevant.

Key numbers

Financial sector work highly automatable (WEF): 32-39%
Work with strong augmentation potential (WEF): 34-37%
Expected financial sector AI investment by 2027: US$97 billion
Corporates citing lack of interoperability as main trade digitalisation barrier: close to 60%
Corporates turning to third-party partners to bridge the gap: around 80%
Corporates seeking advisory support from banking partners: 80%
Increase in AI-enabled attacks in 2025 (CrowdStrike): 89% YoY
Average attacker time from initial access to high-value assets: 29 minutes

Main risks

Concentration risk from reliance on a small number of AI model and compute providers, exposing firms to pricing changes, service availability and vendor roadmap shifts.
Regulatory restrictions on AI capabilities could render critical business processes inoperable or materially more expensive.
AI-enabled fraud and deepfake impersonation have caused billions of dollars in losses in recent years.
AI lowers the barrier to entry for threat actors, who can operate faster and target larger groups.
First documented agentic ransomware shows AI-driven attacks with limited human intervention, adapting as the attack progresses.
Autonomous AI systems may operate outside intended boundaries, as shown by July 2026 OpenAI and Anthropic incidents during cybersecurity evaluations.
Pseudonymity in blockchain wallets creates control gaps: wallet addresses do not prove ownership, key control or continued validity of the control environment.
Fragmented legal, compliance and liquidity processes around tokenised assets may delay institutional scaling.

Corresponding source evidence

As the industry prepares to gather at Sibos 2026, many of these themes will be at the centre of discussions about the future of financial services. · 2026
44Transaction Banking: Bankable Insights Issue 3, 2026 This is why the next phase of banking will not be defined by speed alone, but by orchestration . · 2026
7Transaction Banking: Bankable Insights Issue 3, 2026 Headline act Trust, resilience and interoperability: the new edge As tokenisation, agentic AI and instant rails mature, the technologies themselves will become more widely available. · 2026

Context: the conclusion is Tlines' automated extraction, not the institution's wording. Quoted evidence is publisher wording. Scope and date must travel with the conclusion.

Verify at the original source ↗

Download institution PDF