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 FOOTPRINTsolidASSESSMENT 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

MediaTek Inc

TaiwanEnablersAssignment BorderlineData quality Low–medium

Key messagesolid AI Footprint

MediaTek protects its systems and chips in line with recognised security standards, discloses vulnerabilities regularly and publishes descriptions of its own language models as well as a safety model against fraudulent content. The company has so far only announced its own rules and responsibilities for dealing with artificial intelligence, and in the period examined no court cases, fines or incidents related to artificial intelligence were found.

Opportunities

  • Safetysolid

    Technical security is documented with concrete procedures: ISO 27001 certification, TISAX, ISO/SAE 21434 product certificates, a CISO, an Information Security Committee, red-team exercises, zero trust and around 306 specialists in 2025 (S01, S04), plus a public vulnerability policy with a 90-day …

    Technical security is documented with concrete procedures: ISO 27001 certification, TISAX, ISO/SAE 21434 product certificates, a CISO, an Information Security Committee, red-team exercises, zero trust and around 306 specialists in 2025 (S01, S04), plus a public vulnerability policy with a 90-day target and monthly bulletins giving OEMs at least two months' notice (S09, S10); the independent outside view from UpGuard gives grade A, but only for web and e-mail configuration (S18). AI-specific evidence covers the safety model Breeze Guard with its own benchmark and openly stated limitations (S14) as well as data protection measures of the internal platform DaVinci (S17, C). Limiting factors: Breeze-7B was released without any particular handling of harmful content (S13), abuse monitoring for the publicly usable models is not documented, and chapter 4.6 of the 2025 report remained unread. The chip vulnerability CVE-2025-20435 is recorded as a minor controversy (S25).

  • Transparencysolid

    For its own models there are public model cards with base model, licence and benchmarks (S12, S13) as well as technical reports; the Breeze Guard paper explicitly discloses method, results and limitations (S14).

    For its own models there are public model cards with base model, licence and benchmarks (S12, S13) as well as technical reports; the Breeze Guard paper explicitly discloses method, results and limitations (S14). The sustainability reports state openly that the AI governance framework does not yet exist (S03, S04), and the bulletins disclose vulnerabilities systematically (S10). Disclosure remains limited on training data («extensive corpus», S12), missing sections on limitations and safety in the model cards (S12, S13) and missing labelling or explainability for AI functions of the chips (S08).

Risks

  • Responsibilitylimited

    General governance is documented (Audit Committee with risk oversight, internal audit reporting directly to the board, an independent director with AI expertise since 2025, S05, S07), but AI-specific steering is only announced: the AI governance framework is to be created «in the future», AI …

    General governance is documented (Audit Committee with risk oversight, internal audit reporting directly to the board, an independent director with AI expertise since 2025, S05, S07), but AI-specific steering is only announced: the AI governance framework is to be created «in the future», AI application security is merely discussed in the Information Security Committee, and usage rules for AI tools and an audit mechanism for LLM use are planned measures (S01, S03, S04). A published AI policy, risk classification or approval procedures for in-house and purchased AI (DaVinci, AI Factory, Breeze) are missing (S07, S17). For AI matters, there are therefore only principles and intentions.

  • Risk area

    Security flaws in chips built into billions of end devices

  • Risk area

    Publicly usable in-house language models without documented abuse monitoring

  • Risk area

    Internal generative AI with access to chip design and customer data

  • No serious controversy recorded.

What lies behind the assessment

Research of 3 November 2026
44searches
34pages and documents read
11primary documents
1controversies checked
5open points, named openly
The full research: pillars, evidence, controversies and open points
AI Footprint · Enabler · As of 3 November 2026
MediaTek IncOverall rating: solid

MediaTek documents its information and product security with certifications, a vulnerability policy and monthly security bulletins. The company has so far only announced an AI policy and clear responsibilities for AI risks.

Responsibility

40 %
limitedsolidstrong

General oversight is in place, but AI rules and AI responsibility are only announced.

  • +Independent director with AI expertise since 2025 Source ↗
  • +Information security committee reports annually to the board Source ↗
  • –AI governance framework only announced for the future Source ↗
  • –No published AI policy and no named AI responsibility
  • –No AI-related complaints channel and no AI figures
All sources for this pillar (7)

Safety

40 %
limitedsolidstrong

Certified information and product security, but abuse protection for the models is open.

  • +ISO 27001, TISAX and product certificates under ISO/SAE 21434 Source ↗
  • +Vulnerability policy with a 90-day target and monthly bulletins Source ↗
  • +Safety model Breeze Guard with its own benchmark and open limitations Source ↗
  • –Breeze-7B released without particular handling of harmful content Source ↗
  • –Chip vulnerability CVE-2025-20435 affected encrypted data and wallets Source ↗
All sources for this pillar (13)

Transparency

20 %
limitedsolidstrong

Model cards and an open safety paper, but thin information on data and limitations.

  • +Model cards with base model, licence and benchmarks Source ↗
  • +Breeze Guard paper openly states false alarms and limitations Source ↗
  • +Report states openly that the AI framework is still missing Source ↗
  • –Training data described only as an «extensive corpus» Source ↗
  • –Model cards without sections on limitations and safety Source ↗
All sources for this pillar (10)
What is missing for “strong”

These steps are in the company’s own hands.

1Responsibility
Put an AI policy into force

Adopt the announced AI governance framework, name a responsible function and report on implementation with figures.

2Safety
Document abuse protection for the models

Set and disclose usage rules, abuse monitoring and reporting channels for the publicly usable Breeze models.

3Transparency
Complete the model cards

Describe training data, limitations and known failure patterns in all model cards.

Context: 101 companies assessed

Distribution of overall ratings, no ranking
strong
none yet
solid
← MediaTek Inc is one of 35
limited

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

Controversies
1 1 minor

No deduction for controversies.

Largest risk area
Vulnerabilities in widely used chips

Flaws such as CVE-2025-20435 reach end users through device manufacturers, and the roll-out of fixes depends on those manufacturers.

How thoroughly assessed
30 Evidence

11 primary documents read, 44 searches.

Read the full reasoning and all sources
MediaTek Inc · Full reasoning

Well-documented chip and system security, but AI rules only announced

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

1 · Responsibility

limitedWeight 40%

MediaTek has documented general governance with an Audit Committee and an information security committee, but its AI framework, usage rules and audit mechanism are still only planned measures.

What is documented
  • The Audit Committee oversees risks; internal audit reports directly to the board2 4
  • Five of the nine board members are independent, one of them with recognised AI expertise2 4
  • An Information Security Committee chaired by the Co-COO and the CFO reports annually to the board and discusses the security of AI applications1
  • A confidential reporting channel leads to the Chief Audit Executive; for 2024 MediaTek reports no material legal violations and no fines in two years4 1
  • An AI Training Committee exists in the talent function5
The case against
Intentions only
  • The AI governance framework, revised usage rules for AI and cloud tools and an audit mechanism for AI agents are still only stated as goals5 6 1
  • For the internal AI platform DaVinci, data protection measures are described, but no approval or usage rules7
Missing accountability
  • A published AI policy, a risk classification and a named responsibility for AI risks are missing
  • There is no information on AI training, no external audit of AI processes and no AI-related complaints channel
Sources for Responsibility (7)
  1. MediaTek Inc.: 2024 MediaTek Sustainability Report, English version, chapter 4 (Corporate Governance, chapter PDF) · 3 November 2026 · primary source
  2. MediaTek Inc.: 2025 English Annual Report · 3 November 2026 · primary source
  3. 聯發科技股份有限公司: MediaTek Annual Report 2025 (Chinese version) · 3 November 2026 · primary source
  4. MediaTek Inc.: Corporate Governance (Investor Relations) · 3 November 2026 · primary source
  5. MediaTek Inc.: MediaTek 2025 ESG Report (English) · 3 November 2026 · primary source
  6. MediaTek Inc.: 2024 MediaTek ESG Report (English, full document) · 3 November 2026 · primary source
  7. MediaTek DaVinci (documentation wiki dvcbot.atlassian.net): MediaTek DaVinci FAQ (GAI Service Framework) · 3 November 2026 · secondary source

2 · Safety

solidWeight 40%

MediaTek documents its security organisation with certificates, red-team exercises and a public vulnerability policy, but there is no documented abuse monitoring for the publicly usable models.

What is documented
  • ISO 27001 certification, TISAX and product certificates under ISO/SAE 21434; a zero-trust architecture since 20243 1
  • A CISO, an Information Security Committee, red-team exercises and around 306 information security specialists in 20251 3
  • A public vulnerability policy with a 90-day target and monthly bulletins giving device manufacturers at least two months' notice9 10
  • The outside view from UpGuard gives grade A, but only for web and e-mail configuration8
  • Breeze Guard detects fraud and phishing content in the Taiwanese context and was tested with its own benchmark5 12
  • The internal AI factory runs on premises; logging is switched off for Azure OpenAI6 7
The case against
Gaps in AI abuse protection
  • Neither usage restrictions nor abuse monitoring are documented for the publicly usable Breeze models and the LINE service
  • Breeze-7B was released without particular handling of harmful content4
  • Testing and monitoring of the AI agents used internally are not documented
Incidents and gaps in the evidence
  • The chip vulnerability CVE-2025-20435 allowed encrypted data to be read on certain smartphones; the roll-out of the fix depends on the manufacturers13
  • The scope of the ISO 27001 certificate is not stated; the chapter on information security in the 2025 report remained unread
Sources for Safety (13)
  1. MediaTek Inc.: MediaTek 2025 ESG Report (English) · 3 November 2026 · primary source
  2. MediaTek Inc.: 2024 MediaTek ESG Report (English, full document) · 3 November 2026 · primary source
  3. MediaTek Inc.: 2024 MediaTek Sustainability Report, English version, chapter 4 (Corporate Governance, chapter PDF) · 3 November 2026 · primary source
  4. MediaTek Research (Hugging Face): Model Card Breeze-7B-Instruct-v1_0 · 5 March 2024 · primary source
  5. MediaTek Research / arXiv: Taiwan Safety Benchmark and Breeze Guard: Toward Trustworthy AI for Taiwanese Mandarin (arXiv 2603.07286v1) · 3 November 2026 · primary source
  6. NVIDIA: MediaTek AI Factory (Customer Story) · 11 December 2025 · secondary source
  7. MediaTek DaVinci (documentation wiki dvcbot.atlassian.net): MediaTek DaVinci FAQ (GAI Service Framework) · 3 November 2026 · secondary source
  8. UpGuard: MediaTek Security Rating · 3 November 2026 · primary source
  9. MediaTek Inc.: Vulnerability Disclosure Policy · 3 November 2026 · primary source
  10. MediaTek Inc.: Product Security Bulletin December 2025 · 1 December 2025 · primary source
  11. MediaTek Inc.: Privacy Policy · 22 July 2026 · secondary source
  12. MediaTek Inc. (Tek Talk Blog): MediaTek Research Breeze 3: 讓 AI 聽懂台語、說出台味、守護台灣 · 26 February 2026 · secondary source
  13. IT Brief Australia: Critical MediaTek flaw lets attackers steal phone crypto · 12 March 2026 · secondary source

3 · Transparency

solidWeight 20%

MediaTek publishes model cards and discloses method and limitations in the Breeze Guard paper, but training data and limitations of the other models remain vague.

What is documented
  • The model cards for Llama-Breeze2 and Breeze-7B state base model, architecture, licence and benchmarks6 7
  • The Breeze Guard paper discloses false alarms, the restriction to single conversation turns, synthetic data and an incomplete taxonomy8
  • The sustainability reports state openly that an AI governance framework does not yet exist2 3
  • Monthly product security bulletins systematically disclose vulnerabilities with CVE details4
The case against
Thin model information
  • The training data is described only roughly; there is no transparency report6
  • The model cards for Breeze 2 and Breeze-7B contain no sections on limitations; the blog on Breeze 3 gives no error rates6 7 9
Missing labelling
  • The privacy policy does not mention AI5
  • Labelling and explanations for end users are missing for the AI functions of the chips10
Sources for Transparency (10)
  1. MediaTek Inc.: MediaTek 2025 ESG Report (English) · 3 November 2026 · primary source
  2. MediaTek Inc.: 2024 MediaTek ESG Report (English, full document) · 3 November 2026 · primary source
  3. MediaTek Inc.: 2024 MediaTek Sustainability Report, English version, chapter 4 (Corporate Governance, chapter PDF) · 3 November 2026 · primary source
  4. MediaTek Inc.: Product Security Bulletin December 2025 · 1 December 2025 · primary source
  5. MediaTek Inc.: Privacy Policy · 22 July 2026 · secondary source
  6. MediaTek Research (Hugging Face): Model Card Llama-Breeze2-8B-Instruct · 23 January 2025 · primary source
  7. MediaTek Research (Hugging Face): Model Card Breeze-7B-Instruct-v1_0 · 5 March 2024 · primary source
  8. MediaTek Research / arXiv: Taiwan Safety Benchmark and Breeze Guard: Toward Trustworthy AI for Taiwanese Mandarin (arXiv 2603.07286v1) · 3 November 2026 · primary source
  9. MediaTek Inc. (Tek Talk Blog): MediaTek Research Breeze 3: 讓 AI 聽懂台語、說出台味、守護台灣 · 26 February 2026 · secondary source
  10. MediaTek Inc.: Artificial Intelligence (technology page) · 3 November 2026 · secondary source

4 · Why the harm rule does not apply

The harm rule does not apply: in the period examined there is neither a judgment nor an authority decision nor a fine nor a pattern of serious incidents involving its own AI systems.

5 · Controversies (1)

6 · What remains open

  • Whether an AI policy has since been adopted and who is responsible for AI risks in the group
  • Implementation status of the announced audit mechanism for AI agents and the new usage rules for AI tools
  • Scope of the ISO 27001 certification and incident and data protection figures for 2025
  • Terms of use, abuse monitoring and reporting channels for the Breeze models and the LINE service
  • Chapters on information security, risks and proceedings in the 2025 reports, which were truncated on retrieval

7 · How it was assessed

Cut-off date: 3 November 2026 · 44 searches, 34 fetches.

Primary documents read (11)
  • MediaTek 2025 ESG Report, English (to approx. p. 035, two retrievals)
  • 2025 Sustainability Report, English, FinancialFilings (abridged web version, two retrievals)
  • 2024 MediaTek ESG Report, English (to approx. p. 036)
  • 2024 ESG Report, Chapter 4 Corporate Governance (chapter PDF)
  • 2025 English Annual Report (pp. 1–34)
  • 114年聯發科技年報 (S. 1–27)
  • Corporate Governance page, Vulnerability Disclosure Policy, Product Security Bulletin December 2025, Privacy Policy
  • Model cards Llama-Breeze2-8B-Instruct and Breeze-7B-Instruct, Breeze-Guard paper (arXiv 2603.07286)
  • 股東會議事錄 2026
  • MediaTek v Huawei [2025] EWHC 649 (Pat) (to para. 77)
  • NBIM exclusion list

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