Ranking of 500 AI companies

AI Trustworthiness Ranking 2026

A ranking of leading AI companies based on their security, data privacy, transparency, and public perception.

Average score

500 Companies assessed

36 Countries represented

21 AI categories

Rank	Company	Overall trust score	Security	Transparency	Data privacy	Public perception	Country	Category
1	Google	97	100	100	100	91	US	AI Assistants
2	Krisp	97	100	100	100	91	US	Voice Generation & Conversion
3	Fireflies.ai	96	100	100	100	87	US	Office & Productivity
4	Adobe	94	100	100	100	82	US	Image Generation & Editing
5	Magnific	93	100	100	100	81	ES	Image Generation & Editing
6	Writesonic	93	75	100	100	94	US	Writing and Editing
7	Veryfi	92	100	100	100	78	US	Office & Productivity
8	Salesforce	92	100	67	100	87	US	Business Management
9	Grammarly	92	100	100	88	90	US	Writing and Editing
10	Lovable	92	100	100	100	77	SE	Coding & Development

Showing 1 to 10 of 500 results

Research Advisory Board

The AI Trustworthiness Ranking is supported by an independent advisory board of recognized experts. Their insights help provide additional context on the report's findings and their broader implications.

Dr. Akshika Wijesundara (Dr. Aki) Co-founder of SnapDrum and TAI Labs, CTO of Tilli Kids, and Senior AI Advisor to the UN

Dr. Aki is an AI leader with a PhD in Machine Learning. He is a Google AI Accelerator alumnus and an official training provider within the Anthropic Partner Network. His work sits at the intersection of applied research and production deployment: building production-grade AI systems, and teaching AI to engineers and product managers inside Fortune 100 companies.

Voldemaras Kadys Head of Security and Platform Engineering at Mediatech, the digital publishing house behind Cybernews

Voldemaras Kadys has more than 15 years of experience spanning cybersecurity, software engineering, and IT infrastructure, he advises organizations on cyber risk, resilience, and security strategy. His expertise includes security architecture, cloud and infrastructure security, compliance, risk governance, and executive cybersecurity advisory.

Žilvinas Girėnas Head of Product & Engineering at nexos.ai

Žilvinas Girėnas has been a leader in AI adoption, development, and R&D since 2020, when GPT-3 was released. He has worked across a variety of tech companies, from startups like nexos.ai and Productboard to large corporations, banks, and major insurance projects. With 20 years of experience in the tech industry, his career spans software engineering, tech leadership, and product leadership.

Tomas Čečot Technical SEO Manager at Mediatech, the digital publishing house behind Cybernews

Tomas Čečot is an SEO and Generative Engine Optimization specialist. He develops data-driven strategies for traditional and AI-powered search. As an AI Trustworthiness Research Advisory Board member, Tomas contributes a practitioner's perspective on information quality, source transparency, algorithmic visibility, and the responsible evolution of AI search.

Honoree info

The AI Trustworthiness Ranking includes a badge program that recognizes high-performing companies across the overall ranking, individual trust pillars, and specific AI categories.

AI Trustworthiness Leader – Awarded to companies with an overall score of 75+
Category leaders – Awarded to top-performing companies within each AI category
Pillar Leaders – Awarded to top-performing companies across individual trust pillars (data privacy, security, transparency, and public perception)

If you want to license the badge for public use, please contact our team.

Frequently Asked Questions

What is the AI Trustworthiness Ranking?

The AI Trustworthiness Ranking created by Cybernews is an independent evaluation of 650+ leading consumer-facing AI companies based on publicly verifiable trust signals. Each company receives an Overall Trust Score from 0–100, calculated across four pillars: Public Perception, Data Privacy, Organizational Transparency, and Security. Companies scoring 70+ are designated AI Trustworthiness Leaders 2026.

Why did you create this ranking?

As AI adoption accelerates, users, businesses, and regulators need objective criteria to evaluate which AI companies they can trust. Existing AI rankings focus on performance and capabilities; none of them systematically measure trustworthiness based on publicly verifiable signals. This ranking fills that gap.

What is a trustworthy AI?

Trustworthy AI refers to artificial intelligence systems that are built, deployed, and operated in ways users, regulators, and the public can reasonably rely on. In practice, trustworthy AI means a provider gives clear answers to questions like: What data is collected and used to train the model? Who is behind the company? How is the system secured? How do real users describe their experience? The AI Trustworthiness Ranking measures exactly these elements.

What is ethical AI?

Ethical AI is a disciplinary area that guides how artificial intelligence is responsibly built, trained, released, and used. Its goal is to ensure AI delivers meaningful benefits to people and society while preventing harms such as biased algorithmic outcomes, misuse of personal data, and threats to fundamental rights.

Who is the ranking for?

Businesses evaluating AI vendors, individual users choosing AI tools, journalists and researchers covering the industry, AI companies benchmarking themselves against competitors, and regulators monitoring industry practices.

Where does your data come from?

Our data comes from public sources: privacy policies, security pages, terms of service, and public reviews on Reddit, Trustpilot, and YouTube.

How is the overall trust score calculated?

Each company receives an Overall Trust Score from 0–100, weighted across four pillars: Public Perception (35%), Data Privacy (35%), Security (20%), Organizational Transparency (10%).

Public Perception is based on reviews from Trustpilot and G2.

Data Privacy averages four privacy policy disclosures (data collected, model-training use and opt-out, third-party sharing, retention and deletion), scored Clear / Vague / Missing.

Organizational Transparency averages three Yes / Partial / No checks (legal entity and jurisdiction, verifiable physical address, working non-automated contact channel).

Security is a check on three items — a published bug bounty or vulnerability disclosure program, a current ISO/IEC 27001 or SOC 2 Type II certification (both → 100, one → 50, neither → 0), and a Trust or Security page.

How often is the AI Trustworthiness Ranking updated?

Full re-evaluation is annual. All scores and badges are tagged with the year they were earned (e.g., "2026") and must be re-earned each cycle.

Can companies improve their score? How?

Yes. Because every evaluation pillar is tied to publicly verifiable signals, companies can raise their score by strengthening those signals.

Public Perception: Address recurring complaints on Trustpilot and G2, and improve support responsiveness.

Data Privacy: Rewrite vague policy sections into clear, specific language covering data collected, model-training use and opt-out, third-party sharing, and data retention.

Organizational Transparency: Publish the legal entity and jurisdiction, a verifiable physical address, and a working non-automated contact channel.

Security: Launch a public bug bounty or vulnerability disclosure program, obtain an ISO/IEC 27001 or SOC 2 Type II certification, and create a security or trust page.

All information is provided without a warranty of any kind, express or implied. The methodology used to generate these ratings may not capture all factors relevant to evaluate a company's trustworthiness.