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

SK hynix

South KoreaEnablersAssignment FullData quality Very high

Key messagelimited AI Footprint

SK hynix increasingly uses artificial intelligence in its own administration and manufacturing and describes this mainly in press releases, without publicly available rules, responsibilities or reviews for this use. Independent researchers have also demonstrated a security flaw in the company's memory modules, and a former employee was convicted for taking manufacturing know-how.

Disclosure: the research was supported by an AI model made by Anthropic. SK hynix is connected to Anthropic (e.g. as investor, partner, supplier, customer or through people). The assessment was therefore reviewed by people.

Opportunities

  • In the Globalance AI Footprint, none of the three pillars rates higher than «limited». The company is in the index because of its role in the AI value chain.

Risks

  • Responsibilitylimited

    The 2024 sustainability report documents general compliance and ethics structures with a board committee, reporting channels, case numbers and ISO 37301 certification (S14), but neither there nor in the prospectus or the half-year report was there any board or management responsibility for AI, an …

    The 2024 sustainability report documents general compliance and ethics structures with a board committee, reporting channels, case numbers and ISO 37301 certification (S14), but neither there nor in the prospectus or the half-year report was there any board or management responsibility for AI, an AI policy or an approval procedure for in-house and purchased AI (S01, S04). On the use of AI there are only press descriptions: an internal platform within the internal security network with human-in-the-loop feedback, and a gradual introduction of external AI outside national core technology (S11, S12). This is a recognisable approach without verifiable evidence of implementation. Under V5/V12 the pillar remains “limited”, because neither AI-specific oversight (a) nor a published operational AI rule or review procedure for AI (b) is documented.

  • Safetylimited

    Evidence of information and system protection is limited to secondary reports (TISAX for the domestic sites, S18), a criminal judgment that mentions a USB ban and print logging but finds weaker security at the Shanghai office (S22), and older self-descriptions (S16).

    Evidence of information and system protection is limited to secondary reports (TISAX for the domestic sites, S18), a criminal judgment that mentions a USB ban and print logging but finds weaker security at the Shanghai office (S22), and older self-descriptions (S16). A description of the cybersecurity programme and board oversight in the prospectus could not be retrieved (S01, S02). For the AI control of the factories and the internal generative AI there is no information on testing, validation or monitoring (S13, S11). Independently documented is the Phoenix vulnerability in all 15 tested SK hynix DDR5 modules, with no documented response from the manufacturer (S19, S20).

  • Transparencylimited

    SK hynix describes its internal use of AI (GaiA, agents, roadmap) and the plans for AI-controlled factories only in newsroom and press articles, without information on models, data sources or limitations (S11, S10, S13, S12).

    SK hynix describes its internal use of AI (GaiA, agents, roadmap) and the plans for AI-controlled factories only in newsroom and press articles, without information on models, data sources or limitations (S11, S10, S13, S12). The prospectus treats AI as a driver of demand and names AI-related uncertainties as a risk (S01, S02). Model and system cards are applicable only to a limited extent, as SK hynix does not offer AI models to third parties. No AI-specific disclosure was found in the sustainability report (S14).

  • Risk area

    Leakage of manufacturing know-how through the use of internal and external generative AI

  • No serious controversy recorded.

What lies behind the assessment

Research of 3 November 2026
44searches
27pages and documents read
7primary documents
2controversies checked
6open points, named openly
The full research: pillars, evidence, controversies and open points
AI Footprint · Enabler · As of 3 November 2026
SK hynixOverall rating: limited

SK hynix increasingly uses AI in administration and manufacturing and describes this mainly in press articles. An AI policy, responsibility for AI and information on testing are missing; a flaw in DDR5 memory has been independently documented.

Responsibility

40 %
limitedsolidstrong

General compliance structures are documented, but responsibility for AI is missing.

  • +Board committee reviews the compliance system; reporting channels with case numbers Source ↗
  • +External AI only outside national core technology, according to the CEO Source ↗
  • –No board or management responsibility for AI found
  • –No AI policy and no approval procedure for purchased AI
  • –Use of AI described only in press and newsroom articles Source ↗
All sources for this pillar (10)

Safety

40 %
limitedsolidstrong

Protective measures are known only from secondary sources; a chip flaw is independently documented.

  • +TISAX for all domestic sites, according to the trade press Source ↗
  • +USB ban and logging of printouts, according to a criminal judgment Source ↗
  • –Phoenix flaw in all 15 tested DDR5 modules Source ↗
  • –No information on testing or validation of the factory AI Source ↗
  • –No documented reporting channel for product vulnerabilities
All sources for this pillar (12)

AI resource use

Whole group incl. data centres · reporting year 2025
Disclosure lowDetailsClose
Net zero by 2050, emissions rising (5.98 vs 5.62 million t)
Electricity
–not disclosed
Water consumption
–not disclosed
Renewable electricity
–not disclosed
CO₂e scope 1 and 2
–not disclosed
Target
by
Status 2025
Net-zero emissions
2050
5.98 Mt location-based (2023: 5.62)
Keep scope 1+2 at 2020 level (max 6.17 Mt)
2030
2024: 5.50 Mt market-based
33% renewable electricity (step towards RE100)
2030
32% in Korea (2023: 30%)
Save 600 Mt of water (cumulative since 2020)
2030
303 Mt

Targets and status as reported by the company. According to the company, overseas sites have used 100% renewable power since 2022; there is no scope 3 reduction target.

SK hynix has run its overseas sites on 100% renewable electricity since 2022 and in 2025 cut daily water use at its Korean sites by 170,000 tons. Total electricity and water use and an AI-related share are not verifiably disclosed.

  • 2026 Emissions rose two years running despite RE100 target, to 4.99 million t in 2025
Sources

Supplement without a score: not part of the Globalance AI Footprint.

What is missing for “strong”

These steps are in the company’s own hands.

1Responsibility
Set AI responsibility and policy

SK hynix should name board responsibility for AI and publish an AI policy with an approval procedure for in-house and purchased AI.

2Safety
Disclose reporting channel and AI testing

SK hynix should set up a reporting channel for product vulnerabilities, respond publicly to Phoenix and describe the testing and oversight of its factory AI.

3Transparency
Disclose use of AI in reporting

SK hynix should disclose the AI systems it uses, their providers and their limitations in its sustainability report.

Context: 101 companies assessed

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

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

Controversies
2 2 minor

No deduction for controversies.

Largest risk area
Data leakage as generative AI expands

SK hynix is expanding internal and purchased generative AI without publicly documented rules to protect data classed as national core technology.

How thoroughly assessed
30 Evidence

7 primary documents read, 44 searches.

Read the full reasoning and all sources
SK hynix · Full reasoning

General compliance structures, but no public rules for its own use of AI

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

1 · Responsibility

limitedWeight 40%

SK hynix documents ethics and compliance structures with a board committee and reporting channels, but an AI policy, responsibility for AI and an approval procedure are missing.

What is documented
  • The board's Sustainable Management Committee reviews the materiality analysis and the compliance system; the Audit Committee receives ethics and audit reports1
  • The ethics organisation reports directly to the CEO; in 2023, 283 reports and 61 advice requests were received through open and anonymous channels1
  • The compliance management system has been certified to ISO 37301 since July 2023, including data protection1
  • According to the CEO, external AI services are to be introduced gradually after security and architecture review, and only outside national core technology5
  • According to the newsroom, the internal AI platform GaiA runs within the internal security network, with a human feedback loop6 7
The case against
No AI-specific governance
  • No responsibility for AI was found in the 2024 sustainability report or in the parts of the prospectus and half-year report that were read1 8 4
  • No AI policy, no risk classification and no approval procedure for in-house and purchased AI found
  • No AI-related metrics, training rates or external audits
Evidence from the press only
  • The use of AI is described only in newsroom and press articles, without verifiable evidence of implementation6 5
Sources for Responsibility (10)
  1. SK hynix / Korea Exchange (KIND): SK hynix Sustainability Report 2024 · 23 July 2024 · primary source
  2. SK hynix Newsroom: Strengthening Board-Centred Governance (through-strengthening-governance) · 21 June 2021 · secondary source
  3. Stock Analysis: SK hynix Inc. (SKHY) Company Profile · 3 November 2026 · secondary source
  4. SK hynix Inc. / U.S. SEC: Form 6-K August 2026: translation of the 2026 half-year report (excerpt) · 18 August 2026 · primary source
  5. The Asia Business Daily (Asiae): Following Samsung, SK hynix to Adopt ChatGPT... "AI Utilization Is the Key to Competitiveness" · 12 June 2026 · secondary source
  6. SK hynix Newsroom: Launch of Semiconductor-Specific Generative AI Platform 'GaiA' (ai-platform-gaia-launching-2025) · 14 August 2025 · secondary source
  7. The Asia Business Daily (Asiae): SK hynix Completes Development of In-house Generative AI 'GaiA'... Unveils Three-Year Roadmap · 14 August 2025 · secondary source
  8. SK hynix Inc. / U.S. SEC: Form F-1/A, Amendment No. 2 (Registration No. 333-296987), Preliminary Prospectus · 6 July 2026 · primary source
  9. SK hynix Inc. / U.S. SEC: Form 424B4, final prospectus · 10 July 2026 · primary source
  10. SK hynix Newsroom: SK hynix Named One of the 'World's Most Ethical Companies 2025' (selected-ethical-companies-2025) · 12 March 2025 · secondary source

2 · Safety

limitedWeight 40%

Evidence on information protection comes from the trade press and a criminal judgment; there is no information on testing and oversight of the factory AI and the internal AI.

What is documented
  • According to the trade press, the domestic sites Icheon, Bundang and Cheongju are TISAX certified; the assessment level is not stated4
  • A criminal judgment mentions a USB ban and the logging of printouts5
  • According to the half-year report, the transfer of national core technology abroad requires approval12
  • Since September 2020 SK hynix has not supplied Huawei and listed subsidiaries; an annual licence from the US BIS applies to the China fabs in 202610
  • Business continuity is certified to ISO 223013
The case against
Independently documented vulnerability
  • ETH Zurich researchers demonstrated with Phoenix (CVE-2025-6202) an attack that bypasses the protection in all 15 tested DDR5 modules; existing modules cannot be updated7 8
  • No product security notice, PSIRT or reporting channel from SK hynix is documented7 9
Gaps on AI and documents
  • For the AI control of the factories, aimed at “minimal human intervention”, and for the internal AI agents, information on testing, validation and monitoring is missing1 2
  • The court considered security at the Shanghai office to be less strict5
  • The cybersecurity section of the prospectus could not be read10
Sources for Safety (12)
  1. Cleanroom Technology: SK Hynix to build AI-driven autonomous fab by 2030 · 24 March 2026 · secondary source
  2. SK hynix Newsroom: Launch of Semiconductor-Specific Generative AI Platform 'GaiA' (ai-platform-gaia-launching-2025) · 14 August 2025 · secondary source
  3. SK hynix / Korea Exchange (KIND): SK hynix Sustainability Report 2024 · 23 July 2024 · primary source
  4. EE Times Asia: SK hynix Receives TISAX Certification · 10 February 2025 · secondary source
  5. Tom's Hardware: South Korean engineer smuggled out shopping bags full of secret SK hynix docs to give to Huawei · 9 November 2024 · secondary source
  6. SK hynix Newsroom: 보안을 기업문화로 (filling-the-corporate-culture-with-security) · 13 July 2021 · secondary source
  7. ETH Zurich, Computer Security Group (COMSEC): Phoenix: Rowhammer Attacks on DDR5 with Self-Correcting Synchronization · 15 September 2025 · primary source
  8. Tenable: CVE-2025-6202 · 15 September 2025 · primary source
  9. BleepingComputer: New Phoenix attack bypasses Rowhammer defences in DDR5 memory · 15 September 2025 · secondary source
  10. SK hynix Inc. / U.S. SEC: Form F-1/A, Amendment No. 2 (Registration No. 333-296987), Preliminary Prospectus · 6 July 2026 · primary source
  11. SK hynix Inc. / U.S. SEC: Form 424B4, final prospectus · 10 July 2026 · primary source
  12. SK hynix Inc. / U.S. SEC: Form 6-K August 2026: translation of the 2026 half-year report (excerpt) · 18 August 2026 · primary source

3 · Transparency

limitedWeight 20%

SK hynix describes internal AI and AI-controlled factories in press articles, but discloses neither models nor data sources nor limitations.

What is documented
  • The newsroom describes the GaiA platform with a language model chat for all employees and agents for maintenance, analysis and HR questions1 8
  • The CEO announced to employees that purchased AI assistants from third-party providers were being evaluated2
  • The prospectus lists uncertainties from the use and evolution of AI as a risk factor3 4
  • The plans for largely AI-controlled factories by 2030 are described in the trade press6
The case against
No information on systems and limitations
  • No information on the models used, data sources, limitations or error rates of AI in manufacturing and administration6 1
  • According to the press, internal services run on open models, without names or data information2
  • The 2024 sustainability report contains no chapter on AI; the 2025 and 2026 reports were not found5
Sources for Transparency (8)
  1. SK hynix Newsroom: Launch of Semiconductor-Specific Generative AI Platform 'GaiA' (ai-platform-gaia-launching-2025) · 14 August 2025 · secondary source
  2. The Asia Business Daily (Asiae): Following Samsung, SK hynix to Adopt ChatGPT... "AI Utilization Is the Key to Competitiveness" · 12 June 2026 · secondary source
  3. SK hynix Inc. / U.S. SEC: Form F-1/A, Amendment No. 2 (Registration No. 333-296987), Preliminary Prospectus · 6 July 2026 · primary source
  4. SK hynix Inc. / U.S. SEC: Form 424B4, final prospectus · 10 July 2026 · primary source
  5. SK hynix / Korea Exchange (KIND): SK hynix Sustainability Report 2024 · 23 July 2024 · primary source
  6. Cleanroom Technology: SK Hynix to build AI-driven autonomous fab by 2030 · 24 March 2026 · secondary source
  7. SK hynix Inc. / U.S. SEC: Form 6-K August 2026: translation of the 2026 half-year report (excerpt) · 18 August 2026 · primary source
  8. The Asia Business Daily (Asiae): SK hynix Completes Development of In-house Generative AI 'GaiA'... Unveils Three-Year Roadmap · 14 August 2025 · secondary source

4 · Why the harm rule does not apply

The harm rule does not apply: there is no judgment, regulatory decision or fine for harm caused by its own AI; the 2024 criminal judgment concerns a former employee.

5 · Controversies (2)

6 · What remains open

  • Board and management responsibility for AI and cybersecurity; the relevant sections of the prospectus could not be read
  • AI policy, risk classification and approval procedure for internal and purchased AI
  • Validation and human oversight of AI in manufacturing
  • SK hynix's response to the Phoenix vulnerability and reporting channel for product vulnerabilities
  • Sustainability reports 2025 and 2026 and corporate governance report: not found
  • Complete legal proceedings according to the prospectus (section “Litigation and Regulatory Proceedings”)

7 · How it was assessed

Cut-off date: 3 November 2026 · 44 searches, 27 fetches.

Primary documents read (7)
  • Form F-1/A Amendment No. 2 (S01, two retrievals, truncated) and filing index
  • Form 424B4 (S02, truncated)
  • EDGAR filing index (S03)
  • Form 6-K half-year report 2026 (S04, two retrievals); further 6-K filings dated 2026-08-14 (derivatives loss) and 2026-09-09 (Chongqing) without material content
  • Sustainability report 2024 (S14, two retrievals)
  • Newsroom articles on GaiA (S11), ethics (S17), governance 2021 (S15), safety culture 2021 (S16)
  • ETH Zurich research page Phoenix (S19)

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