Updated Sep 30, 2026
Amazon
Amazon commits to reach net-zero carbon emissions across its global operations by 2040. AWS requires customers making consequential decisions with its AI/ML services to evaluate risks and implement appropriate human oversight, testing, and safeguards.
Closest models
5-way tie
74-78% on 6 of 12 dimensions
- GPT-6 Astra78%
- GPT-5.6 Terra75%
- Claude Opus 574%
- Grok 4.674%
- Kimi K374%
These models are too close to separate: each is within three points of the leader, or would lead if one dimension were left out. For scale, a model answering 50 on every one of these dimensions would score 67%.
Values map
Amazon and GPT-6 Astra
Solid company lines join verified positions. Dashed sections cross unknown axes without assigning a value.
Distribution
Purpose
Worker voice
Equality
Speech
Privacy
Regulation
Global outlook
Social change
Environment
Tech pace
AI autonomy
A higher position means closer to the second endpoint. It does not mean better. The Concord Index measures agreement, not quality or safety.
The score covers 6 of 12 dimensions. The others have no verified public position and are not measured.
Most in sync
Global outlook
Worth a closer look
Regulation
50% match. Inspect the questions and evidence below.
Compare every model.
The range shows the lowest and highest result observed across repeated questions. It is not a confidence interval.
Evidence summary.
Each verified dimension starts with the company position and its citations. Open a dimension only when you want the underlying questions or model answers.
Distribution
Market allocation ↔ Redistribution
Not verified
No company position
No literal excerpt establishes an applicable performance-pay, executive-pay, pay-equity-correction, or broad employee-ownership policy.
Purpose
Shareholder returns ↔ Stakeholder welfare
Not verified
No company position
No excerpt establishes a general shareholder-return priority, customer-and-financial-return framing, or stakeholder decision rule.
Worker voice
Management discretion ↔ Employee influence
Not verified
No company position
The corpus does not establish an applicable Amazon workforce policy on management authority, independent grievances, employee feedback, or collective bargaining.
Equality
Uniform treatment ↔ Targeted measures
Not verified
No company position
The verified quotation does not contain the wording its evidence anchor requires.
“We take a people-centric approach to educating the next generation of AI leaders with programs like the AI & ML Scholarship program and We Power Tech to increase access to hands-on learning, scholarships, and mentorship for underserved or underrepresented in tech.”
Responsible AI – Building AI Responsibly – AWS
Speech
Broad expression ↔ Protective moderation
100%
Company position toward protective moderation
AWS states that its Guardrails can filter harmful content from user interactions with generative-AI applications.
“Guardrails help you define denied topics and content filters to remove undesirable and harmful content from interactions between users and your applications.”
Guardrails for Amazon Bedrock helps implement safeguards customized to your use cases and responsible AI policies (preview) | AWS News Blog
Privacy
Individual privacy ↔ Collective oversight
Not verified
No company position
The supplied excerpts do not establish data minimization, necessary-only retention or deletion, or broad company monitoring or disclosure powers.
Regulation
Voluntary standards ↔ Public regulation
0%
Company position toward public regulation
AWS states that it is pursuing adoption of ISO 42001.
“We are continuing to pursue the adoption of ISO 42001, and look forward to working with customers to do the same.”
ISO 42001: A new foundational global standard to advance responsible AI | Artificial Intelligence
Global outlook
Local priority ↔ International openness
75%
Company position toward international openness
Amazon documents operations and employees across multiple world regions.
“Amazon is a global company with more than 1.5 million full- and part-time employees worldwide and operations in Africa, Asia-Pacific, Europe, Latin America, the Middle East, and North America.”
2024-amazon-sustainability-report.pdf
Social change
Continuity ↔ Experimentation
75%
Company position toward experimentation
Amazon explicitly presents innovation as part of its operating approach.
“At Amazon, we combine data and science with passion and innovation to tackle some of the world’s most urgent environmental and societal challenges.”
2024-amazon-sustainability-report.pdf
Environment
Near-term growth ↔ Ecological restraint
75%
Company position toward ecological restraint
Amazon states a net-zero target covering global operations.
“The Climate Pledge—our goal to reach net-zero carbon emissions across our global operations by 2040”
2024-amazon-sustainability-report.pdf
Tech pace
Rapid advancement ↔ Precaution
75%
Company position toward precaution
AWS requires risk evaluation, testing, human oversight, and safeguards for consequential AI/ML uses.
“If you use the AI/ML Services to make consequential decisions, you must evaluate the potential risks of your use case and implement appropriate human oversight, testing, and other use case-specific safeguards to mitigate such risks.”
AWS Responsible AI Policy
AI autonomy
Delegated autonomy ↔ Human authority
Not verified
No company position
The verified quotation does not contain the wording its evidence anchor requires.
“If you use the AI/ML Services to make consequential decisions, you must evaluate the potential risks of your use case and implement appropriate human oversight, testing, and other use case-specific safeguards to mitigate such risks.”
AWS Responsible AI Policy
Leadership context and reported tensions
Personal opinions are kept separate from the company profile. Reported conduct does not silently replace stated commitments.
AWS promotes strong consumer protections through engagement with lawmakers while also pursuing adoption of the voluntary ISO 42001 AI-management standard. aws.amazon.com aws.amazon.com
Protocol values-2026-09-v8. Panel representative-2026-09-v11. Report version 9f94abc8-2c32-411a-9de7-1979a74b7aea.