Skip to content
GLOBALANCE INTELLIGENCEThe economy ahead.
This Week Index 28/100 Sign in
Money & Investments › GI 100 Index
The Great Value ShiftInvestment ThemesGI 100 IndexGI PortfolioStart Future Check
AI FOOTPRINTlimitedASSESSMENT 2026GLOBALANCEINTELLIGENCE
Globalance AI Footprint

Assessment as of November 2026 · next assessment 2027.

Result of the assessment by the Globalance research team under the published methodology. Not a certification or quality label.

How we assess ›

How this rating is builtVideo · 0:52
GI 100 Index · Profile

ABB Ltd

SwitzerlandEnablersAssignment FullData quality Medium

Key messagelimited AI Footprint

ABB uses artificial intelligence widely in industrial products and in its own operations, and in its code of conduct it commits to safe, lawful and ethical use under human control. What has been confirmed by regulators is mainly its rules for protecting personal data, while no AI policy of its own, no clear oversight by top management and no information on the limits of its AI products were found.

Opportunities

  • Safetysolid

    The data protection framework is officially documented: following an EDPB opinion, the Dutch data protection authority approved ABB's Binding Corporate Rules (S28, S29).

    The data protection framework is officially documented: following an EDPB opinion, the Dutch data protection authority approved ABB's Binding Corporate Rules (S28, S29). For technical system protection, a working vulnerability process is documented (ABB PSIRT, CVE assignment, advisories with firmware updates republished by CISA, S14). Neither piece of evidence is AI-specific; on testing, validation, red-teaming or misuse monitoring of the AI products (e.g. Genix Copilot) there are only company statements such as the human-in-the-loop approach (S01, S15, S16). No AI-related security incident was found in the research period.

Risks

  • Responsibilitylimited

    In its current code of conduct ABB sets out an AI principle (safe, lawful, ethical; responsibility for all decisions; meaningful human control), and its Integrated Report 2025 mentions an “AI council” and around 250 AI projects (S05, S01, S02).

    In its current code of conduct ABB sets out an AI principle (safe, lawful, ethical; responsibility for all decisions; meaningful human control), and its Integrated Report 2025 mentions an “AI council” and around 250 AI projects (S05, S01, S02). No AI policy of its own, no risk classification, no approval procedure for in-house or purchased AI and no responsibility of the board or a committee were found; the Corporate Governance Report 2025 could not be retrieved. The training figures (more than 24,000 employees trained in 2025) are concrete but unverified company statements (S31). An approach is therefore supported only by grade C evidence.

  • Transparencylimited

    ABB discloses that Genix Copilot is based on purchased language models (GPT-4 via Azure OpenAI) and describes the scale of its AI use (S15, S16, S01).

    ABB discloses that Genix Copilot is based on purchased language models (GPT-4 via Azure OpenAI) and describes the scale of its AI use (S15, S16, S01). No information on limits, error rates, explainability or system documentation of the AI products was found (S16). Model cards in the narrow sense do not apply, as ABB has no foundation models of its own. Disclosure remains limited to general company statements (grade C/D).

  • Risk area

    Faulty recommendations or security gaps of AI in critical infrastructure

  • Risk area

    Erroneous output of generative AI in operational decisions (Genix Copilot)

  • No serious controversy recorded.

What lies behind the assessment

Research of 3 November 2026
43searches
34pages and documents read
9primary documents
0controversies checked
6open points, named openly
The full research: pillars, evidence, controversies and open points
AI Footprint · Enabler · As of 3 November 2026
ABB LtdOverall rating: limited

In its code of conduct ABB commits to safe, lawful and ethical use of AI under human control. What regulators have confirmed is mainly data protection; an AI policy and board oversight are missing.

Responsibility

40 %
limitedsolidstrong

An AI principle is in the code, but a policy and top-level oversight are missing.

  • +Code of conduct requires meaningful human control over AI Source ↗
  • +An “AI council” coordinates around 250 AI projects Source ↗
  • –No documented responsibility of the board or a committee for AI
  • –No AI policy, risk classification or approval procedure
  • –Training figures are unverified company statements Source ↗
All sources for this pillar (7)

Safety

40 %
limitedsolidstrong

Data protection and handling of vulnerabilities are officially documented, but not AI-specific.

  • +Binding Corporate Rules approved by the Dutch authority Source ↗
  • +Vulnerabilities are disclosed and fixed with updates Source ↗
  • –No AI testing, validation or red-teaming documented
  • –No misuse monitoring and no AI incident process found
All sources for this pillar (11)

Transparency

20 %
limitedsolidstrong

ABB quantifies the scale of its AI use but states no limits of its products.

  • +Scale of AI use quantified in the Integrated Report Source ↗
  • –No information on limits or error rates of the AI products
  • –No labelling rules for AI-generated output found
  • –No system documentation for Genix Copilot and the Genix AI Suite
All sources for this pillar (6)
What is missing for “strong”

These steps are in the company’s own hands.

1Responsibility
Set an AI policy and oversight

Publish an AI policy of its own with risk classification and disclose the responsibility of the board or a committee for AI.

2Safety
Test AI products and disclose the tests

Document testing, validation and red-teaming for Genix Copilot and AI in control systems and set up a reporting process for AI malfunctions.

3Transparency
Document the limits of AI products

Publish system documentation with conditions of use, limits and labelling of AI-generated output.

Context: 101 companies assessed

Distribution of overall ratings, no ranking
strong
none yet
solid
limited
← ABB Ltd is one of 66

No company reaches “strong” today. Disclosure on how companies handle AI is still young across the market.

Controversies
0 none found

No deduction for controversies.

Largest risk area
AI errors in critical infrastructure

No tests, validation or red-teaming results have been disclosed for AI in control and analytics systems for energy, utilities and the process industry.

How thoroughly assessed
24 Evidence

9 primary documents read, 43 searches.

Read the full reasoning and all sources
ABB Ltd · Full reasoning

AI principle in the code, but neither an AI policy nor documented top-level oversight

Overall rating “limited”. No deduction for controversies. The superscript numbers link directly to the source.

1 · Responsibility

limitedWeight 40%

ABB bases responsibility on a clause in its code of conduct and an “AI council”, not on an AI policy.

What is documented
  • The current code of conduct requires safe, lawful and ethical use of AI, responsibility for all decisions and meaningful human control4
  • The 2020 version of the code had no AI passage; the clause is a later addition5
  • The Integrated Report 2025 mentions an “AI council” for sharing best practice and around 250 ongoing AI projects1 2
  • The code of conduct expressly also applies to the board; a Speak Up Helpline with a ban on retaliation is in place3 7
  • ABB states that it reached more than 24,000 employees with AI training in 2025, including on safe and responsible use6
The case against
Governance and oversight
  • No AI policy of its own, no risk classification and no approval procedure for in-house or purchased AI were found
  • Responsibility of the board or a committee for AI is not documented; the Corporate Governance Report 2025 could not be retrieved
  • The mandate, composition and reporting line of the “AI council” are not described1
Evidence
  • There are no AI metrics, no external audit or certification and no AI-related complaint channels with documented remedies
  • The training figures are unverified company statements6
Sources for Responsibility (7)
  1. ABB Ltd: ABB Integrated Report 2025 · 19 February 2026 · secondary source
  2. ABB Ltd: ABB Integrated Report 2025 (second storage URL) · 19 February 2026 · secondary source
  3. ABB Ltd: ABB Code of Conduct (German landing page) · 3 November 2026 · secondary source
  4. ABB Ltd: ABB Code of Conduct – Protecting assets and reputation (German version) · 3 November 2026 · secondary source
  5. ABB Ltd (abgelegt im Lobbyregister des Deutschen Bundestages): ABB Code of Conduct, Rev 0 (June 2020) · 1 June 2020 · secondary source
  6. ABB Ltd: How ABB employees use generative AI in the workplace: Training, Tools, and Impact · 22 April 2026 · secondary source
  7. ABB Ltd: ABB Code of Conduct (English landing page) · 3 November 2026 · secondary source

2 · Safety

solidWeight 40%

The security evidence is official but concerns data protection and vulnerabilities in general, not the AI products.

What is documented
  • In 2026, following an EDPB opinion, the Dutch data protection authority approved the ABB Group's Binding Corporate Rules7 8
  • According to ABB, data subjects can complain via a contact form, email and postal address9
  • In 2026 ABB PSIRT published advisories on three vulnerabilities in AWIN gateways with firmware updates; CISA republished them4
  • Cyber risk, including AI-enabled threats, is listed as a top risk in risk management1
  • The code of conduct requires compromised personal data to be reported to the IS service desk immediately10
The case against
Testing of the AI products
  • On testing, validation and red-teaming of Genix Copilot and other AI products there are only company statements such as the human-in-the-loop approach1
  • The ISA/IEC 62443-4-1 certification covers only one business line, dates from 2023 and is not AI-specific3
Misuse and incidents
  • No usage rules for AI products, no misuse monitoring and no reporting process for AI malfunctions were found
  • Rules on handling personal data in the internal tool “Abby” are not described11
Sources for Safety (11)
  1. ABB Ltd: ABB Integrated Report 2025 · 19 February 2026 · secondary source
  2. ABB Ltd: ABB recognized as a Leader in 2025 Verdantix Green Quadrant for Industrial AI Analytics · 3 November 2026 · secondary source
  3. ABB Ltd: ISA/IEC 62443 cyber security standard confirms protection integral to development for ABB Electrification Smart Power · 1 August 2023 · secondary source
  4. CISA (republication of an ABB CSAF advisory) via GCVE database: ICSA-26-120-05: ABB AWIN Gateways (ABB PSIRT 4JNO000329) · 13 March 2026 · primary source
  5. ABB Ltd: Digitalization and AI – A customer-focused approach (website) · 3 November 2026 · secondary source
  6. ABB Ltd / U.S. Securities and Exchange Commission (EDGAR): ABB Ltd Form 20-F for fiscal year 2023 · 3 November 2026 · primary source
  7. European Data Protection Board (EDPB): Opinion 4/2026 on the draft decision of the Dutch Supervisory Authority regarding the Controller Binding Corporate Rules of the ABB Group · 15 January 2026 · primary source
  8. Autoriteit Persoonsgegevens (Dutch data protection authority): Decision approving Controller Binding Corporate Rules of ABB · 12 February 2026 · primary source
  9. ABB Ltd: ABB Binding Corporate Rules (website) · 1 January 2026 · secondary source
  10. ABB Ltd: ABB Code of Conduct – Protecting assets and reputation (German version) · 3 November 2026 · secondary source
  11. ABB Ltd: How ABB employees use generative AI in the workplace: Training, Tools, and Impact · 22 April 2026 · secondary source

3 · Transparency

limitedWeight 20%

ABB discloses the scale of its AI use, but no limits, error rates or system documentation for its AI products.

What is documented
  • The Integrated Report 2025 puts AI use at around 250 ongoing AI projects3 4
  • The Form 20-F for 2023 describes the Genix Industrial Analytics and AI Suite as an offering5
  • According to ABB, the internal tool “Abby” works with more than 15 language models and has more than 35,000 users6
The case against
Limits and documentation
  • No information on the limits, error rates and explainability of the AI products was found
  • Labelling rules for AI-generated output and system documentation for Genix Copilot are missing
  • Model cards do not apply because ABB offers no foundation models of its own; data information for its own analytics models is missing
Sources for Transparency (6)
  1. ABB Ltd: Helping industries do better with generative AI: ABB launches Genix Copilot with Microsoft · 3 November 2026 · secondary source
  2. ABB Ltd: ABB Genix Copilot (product page) · 3 November 2026 · secondary source
  3. ABB Ltd: ABB Integrated Report 2025 · 19 February 2026 · secondary source
  4. ABB Ltd: ABB Integrated Report 2025 (second storage URL) · 19 February 2026 · secondary source
  5. ABB Ltd / U.S. Securities and Exchange Commission (EDGAR): ABB Ltd Form 20-F for fiscal year 2023 · 3 November 2026 · primary source
  6. ABB Ltd: How ABB employees use generative AI in the workplace: Training, Tools, and Impact · 22 April 2026 · secondary source

4 · Why the harm rule does not apply

The harm rule does not apply: no AI-related ruling, fine or pattern of serious incidents was found in the 36-month window. The unresolved allegation about the use of ABB tools in Russia has no AI link and does not count as a controversy.

5 · Controversies (0)

No AI controversy found.

6 · What remains open

  • Responsibility of the board or a committee for AI and cyber security; the Corporate Governance Report 2025 could not be retrieved
  • The full text of the Sustainability Statement 2025 and the governance chapter of the Integrated Report 2025 were not read
  • Testing, validation and red-teaming of the AI products and documented limits of Genix Copilot
  • Completion status of the sale of the robotics division to SoftBank as of the reference date
  • Outcome of the internal review of the allegation that ABB tools are used in Russia to produce weapons
  • Version date of the current code of conduct with the AI clause

7 · How it was assessed

Cut-off date: 3 November 2026 · 43 searches, 34 fetches.

Primary documents read (9)
  • ABB Integrated Report 2025 (two filing URLs, to approx. p. 55)
  • ABB Sustainability Statement 2024 (to approx. p. 39)
  • ABB Ltd Form 20-F FY2023 (to Item 4)
  • ABB Code of Conduct, current web version (English/German) incl. section on Artificial Intelligence
  • ABB Code of Conduct Rev 0 (2020)
  • EDPB Opinion 4/2026 on the ABB BCR-C
  • Decision of the Autoriteit Persoonsgegevens of 2026-02-12 on the ABB BCR-C
  • CISA ICS Advisory ICSA-26-120-05 (ABB PSIRT)
  • ABB press releases on Genix Copilot, AGM 2025, Annual Reporting Suite 2025, Verdantix

Research of 3 November 2026 under Globalance AI Footprint rulebook version 0

We assess documented practice in dealing with AI; this is not an investment recommendation. Three pillars, weights 40/40/20, harm rule.

Research of 3 November 2026 under Globalance AI Footprint rulebook version 0. Documented observations from public sources, no investment recommendation.

Important Notice. This material is provided for information purposes only and does not constitute an offer or solicitation to buy or sell any financial instrument, nor does it constitute investment advice or a recommendation. The GI 100 Index is a rules-based strategy index of Globalance Bank. Investments involve risk, including the loss of capital invested. The final product documents (including FinSA) shall prevail in all cases.

Your company? Information on inclusion in the GI 100 Index →

What comes next