Developer docs
API playgroundTry for free, no card

Search company profiles

Dexai

Full company profile

uuid00qkhdg

Namestring
Dexai
Legal namestring
Dexai S.r.l.
Websiteurl
dexai.eu
Company typeenum
Private
Founded yearint
2021
Descriptiontext

Dexai S.r.l. is a Rome-based, fully-remote AI ethics, governance, and compliance consultancy founded in 2021 by Federico Sartore (Head of Legal) and Piercosma Bisconti (Head of Scientific Research). It sells four bespoke professional-service lines — AI Models Evaluation (LLM benchmarking, robustness/adversarial testing, fairness/bias analysis, explainability), AI Literacy (training tied to the EU AI Act's Article 4 obligation), AI Ethics (ethics-by-design advisory, stakeholder engagement), and AI Act Compliance (risk classification, gap analysis, conformity assessment support) — typically scoped and priced per engagement via inbound contact at [email protected].

Its core technology is a set of proprietary AI evaluation methodologies aligned with ISO and CEN-CENELEC trustworthy AI standards and the EU AI Act, delivered by practitioners actively conducting frontier research. The flagship research initiative Icaro Lab, a collaboration with Sapienza University of Rome and Sant'Anna School of Advanced Studies, has produced high-profile safety research (poetic-jailbreak study with 62% attack success across 25 models from nine vendors including ChatGPT, Claude, and Gemini). Co-founder Piercosma Bisconti also leads the CEN-CENELEC European 'AI Trustworthiness Framework' standard.

Commercially, Dexai operates a B2B, sales-led, relationship-driven GTM targeting EU enterprises, public authorities, and standardization bodies needing AI Act compliance. Revenue is generated through three streams: bespoke consulting, training/workshop delivery, and co-funding from six EU-funded Horizon Europe and LIFE projects (ENGAGE4ESG, SOLARIS, LUCIA, AI4HOPE, CERTAIN, SHIELD). Pricing is quote-based and not publicly disclosed; the company remains privately held and bootstrapped with no external funding rounds disclosed.

Short descriptiontext

Dexai is a Rome-based, fully-remote AI ethics consultancy founded in 2021 that helps EU enterprises, public authorities, and standardization bodies evaluate LLMs, train staff, and comply with the EU AI Act via four bespoke service lines (Model Evaluation, AI Literacy, AI Ethics, AI Act Compliance) and frontier safety research via Icaro Lab.

Operating statusenum
Operating
Ownership categoryenum
Headcount rangeband
1–10
akta.pro rankint
HeadquartersRome, Italy
HQ citystring
Rome
HQ countrystring
Italy
HQ regionstring
Europe
Markets served

Serves global market

Offices1 record

Each record includes

City, Country, Type, Description, Source

Keyword5 values
AI ethics consulting, AI compliance services, LLM evaluation, AI governance advisory, EU AI Act compliance
Industry5 codes
1Responsible AI, AI Governance & Compliance Services
CodeBPAEAHAJPrimaryYes
2Regulatory Readiness & Audit Automation (e.g., EU AI Act, NIST AI RMF, ISO/IEC 42001)
CodeHDAAAMAEPrimaryNo
3AI Governance, Risk & Compliance (GRC) Platforms
CodeHDAAAMAAPrimaryNo
4Model Transparency, Explainability & Interpretability (XAI)
CodeHDAAAMABPrimaryNo
5Third-Party Model/Vendor Risk & Supply-Chain Assurance
CodeHDAAAMAKPrimaryNo
NAICS code1 code
  • Research and Development in the Social Sciences and Humanities541720
SIC code2 codes
  • Services-Engineering, Accounting, Research, Management8700
  • Services-Management Services8741
Product category
AI Governance & Compliance Consulting
GTM motion4 records

Each record includes

Type, Description, Source

Revenue model3 records
1Professional Consulting Services
TypeProfessional Services
Description

Bespoke, tailored engagements across four service lines: AI Models Evaluation, AI Literacy training, AI Ethics advisory, and AI Act Compliance. Each engagement is scoped to the client's specific AI systems, business context, and regulatory risk level.

dexai.eu
2EU-Funded Research Project Participation
TypeProfessional Services
Description

Participation in EU-funded Horizon Europe and LIFE programme projects (ENGAGE4ESG, SOLARIS, LUCIA, AI4HOPE, CERTAIN, SHIELD) providing co-funding and project-based revenue for ethics, governance, and compliance work-packages.

dexai.eu
3Training & Workshop Delivery
TypeProfessional Services
Description

Delivery of AI literacy workshops, executive briefings, ethics-by-design workshops, and use-case walkthroughs as part of the AI Literacy and AI Ethics service lines.

dexai.eu
Marketing channels8 records

Each record includes

Title, Type, Stage, Description, Source

Distribution channels3 records

Each record includes

Title, Type, Scope, Target buyer, Description, Source

Cost components4 values
Personnel, Technology or R&D, Marketing or Sales, Operations
Pricing details1 tier
1Bespoke consulting engagements
ModelOtherBilling cadenceMulti-year contract
Notes

Pricing is quote-based and tailored to each client's unique requirements; not publicly disclosed.

dexai.eu
GTM typeB2B
B2B
Offering typeServices
Services
Core offering1 text field

Dexai is an AI ethics and governance consultancy that delivers bespoke advisory engagements across four service lines: AI Models Evaluation (independent LLM benchmarking, robustness, fairness, and explainability assessments), AI Literacy (training programs tied to the EU AI Act Article 4 obligation), AI Ethics (ethics-by-design frameworks and stakeholder engagement), and AI Act Compliance (risk classification, gap analysis, conformity documentation, and audit readiness). Each engagement is scoped to the client's AI systems, business context, and regulatory exposure.

Differentiator
Functional benefit
Problem solved
Quantifiable outcome1 of 3 values shown
  • 62% attack success rate identified using handcrafted poetic prompts across 25 AI models
+2 more records
Product overview1 text field

Dexai operates as a consulting and advisory firm rather than a unified software platform, offering four core services that are typically sold as standalone engagements but commonly combined across the AI lifecycle: AI Models Evaluation (technical assessment of LLMs and other AI systems), AI Literacy (training programs tied to the EU AI Act's literacy obligation), AI Ethics (ethics-by-design and governance consulting), and AI Act Compliance (risk classification, gap analysis, and conformity support under the EU AI Act). Underpinning these services is Dexai's research initiative Icaro Lab, which investigates AI safety vulnerabilities such as poetry-based jailbreaks across state-of-the-art LLMs, and Dexai's participation in EU-funded projects (e.g., ENGAGE4ESG, SOLARIS, LUCIA, AI4HOPE, CERTAIN, SHIELD) where it contributes AI ethics, governance, and regulatory expertise.

Product and service5 records
1AI Models Evaluation
CategoryAI Model Evaluation
Description

Independent, rigorous evaluation of AI models (especially large language models) covering performance benchmarking, robustness testing under perturbations and adversarial inputs, fairness and bias analysis, explainability/transparency checks, and qualitative and quantitative risk assessment aligned with the client's regulatory context.

2AI Literacy
CategoryAI Training & Education
Description

Tailored training and education programs that build foundational AI awareness, critical thinking, and practical understanding across executives, non-technical staff, and policy/legal teams, delivered via custom learning paths, workshops, executive briefings, use-case walkthroughs, and ongoing learning support.

3AI Ethics
CategoryAI Ethics Consulting
Description

Consulting support for organizations to integrate ethical principles (transparency, fairness, accountability, human oversight) into the design, development, and deployment of AI systems, including ethics assessments, internal framework development, ethics-by-design workshops, stakeholder engagement guidance, and ongoing advisory support for ethics boards and review committees.

4AI Act Compliance
CategoryRegulatory Compliance Consulting
Description

Strategic and operational guidance to help organizations comply with the EU AI Act, including risk classification of AI systems, gap analysis against regulatory requirements, tailored compliance roadmap, technical documentation support for conformity and audit readiness, and ongoing advisory on regulatory developments.

5Icaro Lab
CategoryAI Safety Research
Description

Dexai's research initiative on AI safety and trustworthiness. Icaro Lab studies how poetic language can bypass AI safety guardrails in large language models and exposes blind spots in current AI safety testing methodologies, contributing to the development of more robust evaluation and red-teaming practices.

Scale indicator8 records

Each record includes

Type, Value, Description, Source

Partnership19 partners
Strategic tierFlagship / Core Research PartnershipTypeStrategic or Co-development Partner
Description

Icaro Lab is a research collaboration between Sapienza University of Rome, DexAI, and Sant'Anna School of Advanced Studies. It produced the high-profile poetic jailbreak study that achieved 62% attack success rate across 25 AI models.

Strategic tierCore Research PartnerTypeStrategic or Co-development Partner
Description

Partner in the Icaro Lab collaboration. Sapienza is also displayed on the Dexai website partner logo wall. The university co-develops research with Dexai on AI safety and ethics.

Strategic tierCore Research PartnerTypeStrategic or Co-development Partner
Description

Partner in the Icaro Lab collaboration with Sapienza University of Rome and DexAI. The collaboration produced research on bypassing AI safety guardrails via poetic language.

4ENGAGE4ESG (LIFE programme)
Strategic tierCore Eu ProjectTypeStrategic or Co-development Partner
Description

LIFE programme project aligning real estate lending with EU climate goals and ESG standards. Dexai supports development of compliant, ethical frameworks for data use and sustainability reporting.

dexai.eu
Strategic tierCore Eu ProjectTypeStrategic or Co-development Partner
Description

Horizon Europe project combatting disinformation and echo chambers using generative AI. Dexai contributes to the ethical design and stakeholder engagement strategy.

6LUCIA (Horizon Europe)
Strategic tierCore Eu ProjectTypeStrategic or Co-development Partner
Description

Horizon Europe project aimed at reducing cancer inequalities across Europe by developing AI-driven solutions for early detection and prevention. Dexai focuses on AI governance, ethics, and stakeholder coordination.

dexai.eu
7AI4HOPE
Strategic tierCore Eu ProjectTypeStrategic or Co-development Partner
Description

Project exploring AI-driven support tools for palliative care in dementia. Dexai leads activities related to legal compliance, auditing practices, and inclusive AI system design in sensitive healthcare environments.

dexai.eu
8CERTAIN (Horizon Europe)
Strategic tierCore Eu ProjectTypeStrategic or Co-development Partner
Description

Horizon Europe initiative working on transparency and certification of trustworthy AI systems. Dexai helps define ethical and regulatory requirements and supports creation of practical tools for responsible AI lifecycle management.

dexai.eu
Strategic tierCore Eu ProjectTypeStrategic or Co-development Partner
Description

Innovation Action under Horizon Europe (2025–2029), SHIELD focuses on early detection of pancreatic cancer in individuals with genetic predispositions, validating a new blood-based diagnostic test. Dexai contributes to ethical, legal, and social framework of the study and supports patient engagement and clinical translation.

Strategic tierSupporting Partner Via Co-Founder AffiliationTypeImplementation/ SI/ Consulting Partner
Description

Co-founder Federico Sartore is a Partner at PANETTA Law Firm, advising companies on data protection, cybersecurity, and AI governance. The firm supports Dexai's legal and ethical advisory work.

11ANSA
Strategic tierMinor — Logo PartnerTypeOthers
Description

Listed on Dexai's 'Our partners' logo wall on the homepage. Specific collaboration details not disclosed in source.

dexai.eu
Strategic tierMinor — Logo PartnerTypeOthers
Description

Listed on Dexai's 'Our partners' logo wall on the homepage. Likely the University of Maribor or related institution. Specific collaboration details not disclosed.

Strategic tierMinor — Logo PartnerTypeOthers
Description

Listed on Dexai's 'Our partners' logo wall on the homepage. Likely the University of Amsterdam or Amsterdam-based institution. Specific collaboration details not disclosed.

Strategic tierMinor — Logo PartnerTypeOthers
Description

Libera Università Maria Ss. Assunta (LUMSA) — listed on Dexai's 'Our partners' logo wall. Specific collaboration details not disclosed.

Strategic tierSupporting PartnerTypeStrategic or Co-development Partner
Description

Spanish applied research center listed on Dexai's partner logo wall. Likely a research consortium partner in EU projects.

16RED (red_r)
Strategic tierMinor — Logo PartnerTypeOthers
Description

Listed on Dexai's 'Our partners' logo wall. Specific collaboration details not disclosed in source.

dexai.eu
Strategic tierMinor — Logo PartnerTypeOthers
Description

Listed on Dexai's 'Our partners' logo wall. Specific collaboration details not disclosed in source.

Strategic tierSupporting PartnerTypeStrategic or Co-development Partner
Description

Listed on Dexai's partner logo wall — likely Humanitas Research Hospital or Humanitas University, an Italian healthcare partner for healthcare AI projects (e.g., AI4HOPE).

Strategic tierSupporting PartnerTypeStrategic or Co-development Partner
Description

Italian national inter-university consortium for informatics listed on Dexai's partner logo wall — likely a research and standardization partner.

Recent move6 records

Each record includes

Date, Type, Title, Description, Source

Expansion highlight5 records

Each record includes

Type, Description

Peers10 records
TypeDirect peer
Description

Trustible is an AI governance and policy management platform that helps enterprises operationalize responsible AI programs and map to regulations including the EU AI Act. Directly comparable buyer, problem space, and EU regulatory framing as Dexai.

TypeEmerging player
Description

Fiddler AI provides an AI observability and explainability platform with bias monitoring and model performance tracking. Adjacent to Dexai's model evaluation and explainability/transparency work, but more productized and focused on continuous monitoring rather than conformity assessments.

TypeBroad incumbent
Description

Deloitte's Trustworthy AI practice offers enterprise-scale AI governance, EU AI Act readiness, model risk management, and responsible AI advisory. Represents the broad-incumbent competitor with deep enterprise relationships, large bench, and tool-enabled delivery that Dexai ultimately competes against for AI Act advisory budgets.

TypeDirect peer
Description

Credo AI provides an AI governance platform covering AI risk management, regulatory mapping (including EU AI Act and NIST AI RMF), and audit-ready documentation. Targets the same enterprise governance, risk, and compliance buyer as Dexai and competes for AI Act compliance budgets; differentiates via SaaS platform vs. Dexai's bespoke services.

TypeDirect peer
Description

Fairly AI provides AI risk and compliance management software with EU AI Act assessments, model evaluation, and governance workflows. Comparable to Dexai's AI Model Evaluation and AI Act Compliance services but delivered via a SaaS platform.

TypeEmerging player
Description

Enzai offers an AI governance and risk management platform focused on helping organizations comply with the EU AI Act and emerging AI regulations. Closely comparable mission and EU regulatory framing as Dexai; smaller scale and more product-focused.

TypeDirect peer
Description

Saidot is a Finnish AI governance SaaS focused on EU AI Act transparency, AI literacy, and stakeholder engagement tooling. Directly comparable to Dexai on AI Act compliance and AI literacy, with similar European policy-shaping positioning; differs in being more productized.

TypeDirect peer
Description

Holistic AI is an AI governance, risk, and compliance platform that explicitly helps organizations operationalize the EU AI Act — offering AI risk assessment, bias audits, and regulatory readiness tooling. Closely comparable to Dexai in serving the same enterprise buyer navigating AI Act conformity, though Holistic AI is more product-led whereas Dexai is consulting-led.

TypeDirect peer
Description

Monitaur is an AI governance, risk, and compliance platform with model monitoring, bias testing, and audit-ready documentation. Overlaps with Dexai's model evaluation and governance framework services, especially for regulated US enterprises; serves a similar GRC buyer.

TypeEmerging player
Description

Responsible AI Institute is a non-profit developing AI assurance and certification frameworks, including assessment against the EU AI Act. Comparable to Dexai's standards and conformity work, with broader certification ecosystem scope rather than consulting delivery.

Market position
Strengths4 records

Each record includes

Headline, Details, Source

Weaknesses4 records

Each record includes

Headline, Details, Source

Competitive moat4 records

Each record includes

Type, Details

Key risks6 records

Each record includes

Headline, Details, Source

Key highlights6 records

Each record includes

Headline, Details, Source

Customer concentration

Classification, Details

Segment6 records

Each record includes

Title, Type, Primary, Description, Pain point addressed, Use case, Source

Ideal customer profile4 records

Each record includes

Profile, Firmographic size, Sales motion, Sales cycle length, Buying structure, Purchase trigger, Buyer persona, Geography, Industry vertical, Primary use case, Description, Pain points, Evidence proof points, Target buyer

Technology focused
Yes
API detail
Has APIbool
No

Docs URL, Description

AI capability7 records

Each record includes

Type, Description, Source

AI maturity
App detail

Has app

Feature7 records

Each record includes

Title, Differentiator, Description, Source

Core technology
Revenue estimate
Valuation estimate
Number of profiles
Profiles11 records

Each record includes

Name, Designation, Designation category, Overview, Profile commentary, Source

No data
Compliance2 records

Each record includes

Name, Class, Description

Funding overview

Funding stage, Last funding date, Total funding USD

Funding rounds

Each record includes

Round, Amount USD, Date, Pre money valuation, Total investors, Investors, News

Investors

Each record includes

Name, Type, Date of entry, Rounds participated, Website

Funding detail is available on the Subscription and Enterprise plan.Contact sales →

M&A

Each record includes

Name, Acquisition type, Announced date, Completed date, Status, Website, News

Investment

Each record includes

Name, Round, Announced date, Lead investor, Website, News

M&A and investment is available on the Subscription and Enterprise plan.Contact sales →

Dexai

AI Governance & Compliance Consultingdexai.eu

Dexai is a Rome-based, fully-remote AI ethics consultancy founded in 2021 that helps EU enterprises, public authorities, and standardization bodies evaluate LLMs, train staff, and comply with the EU AI Act via four bespoke service lines (Model Evaluation, AI Literacy, AI Ethics, AI Act Compliance) and frontier safety research via Icaro Lab.

What Dexai does

Dexai S.r.l. is a Rome-based, fully-remote AI ethics, governance, and compliance consultancy founded in 2021 by Federico Sartore (Head of Legal) and Piercosma Bisconti (Head of Scientific Research). It sells four bespoke professional-service lines — AI Models Evaluation (LLM benchmarking, robustness/adversarial testing, fairness/bias analysis, explainability), AI Literacy (training tied to the EU AI Act's Article 4 obligation), AI Ethics (ethics-by-design advisory, stakeholder engagement), and AI Act Compliance (risk classification, gap analysis, conformity assessment support) — typically scoped and priced per engagement via inbound contact at [email protected].

Its core technology is a set of proprietary AI evaluation methodologies aligned with ISO and CEN-CENELEC trustworthy AI standards and the EU AI Act, delivered by practitioners actively conducting frontier research. The flagship research initiative Icaro Lab, a collaboration with Sapienza University of Rome and Sant'Anna School of Advanced Studies, has produced high-profile safety research (poetic-jailbreak study with 62% attack success across 25 models from nine vendors including ChatGPT, Claude, and Gemini). Co-founder Piercosma Bisconti also leads the CEN-CENELEC European 'AI Trustworthiness Framework' standard.

Commercially, Dexai operates a B2B, sales-led, relationship-driven GTM targeting EU enterprises, public authorities, and standardization bodies needing AI Act compliance. Revenue is generated through three streams: bespoke consulting, training/workshop delivery, and co-funding from six EU-funded Horizon Europe and LIFE projects (ENGAGE4ESG, SOLARIS, LUCIA, AI4HOPE, CERTAIN, SHIELD). Pricing is quote-based and not publicly disclosed; the company remains privately held and bootstrapped with no external funding rounds disclosed.

Dexai firmographics

Firmographics
Name
Dexai
Legal name
Dexai S.r.l.
Website
https://dexai.eu
Company type
Private
Founded year
2021
Operating status
Operating
Headcount range
1–10 employees
Short description
Dexai is a Rome-based, fully-remote AI ethics consultancy founded in 2021 that helps EU enterprises, public authorities, and standardization bodies evaluate LLMs, train staff, and comply with the EU AI Act via four bespoke service lines (Model Evaluation, AI Literacy, AI Ethics, AI Act Compliance) and frontier safety research via Icaro Lab.
Ownership category
akta.pro rank

Dexai industry classification

Industry
Product category
AI Governance & Compliance Consulting
NAICS
Research and Development in the Social Sciences and Humanities (541720)
SIC
Services-Engineering, Accounting, Research, Management (8700), Services-Management Services (8741)
akta.pro primary industry
Responsible AI, AI Governance & Compliance Services (BPAEAHAJ)
akta.pro secondary industries
Regulatory Readiness & Audit Automation (e.g., EU AI Act, NIST AI RMF, ISO/IEC 42001) (HDAAAMAE), AI Governance, Risk & Compliance (GRC) Platforms (HDAAAMAA), Model Transparency, Explainability & Interpretability (XAI) (HDAAAMAB), Third-Party Model/Vendor Risk & Supply-Chain Assurance (HDAAAMAK)

Keywords

  • AI ethics consulting
  • AI compliance services
  • LLM evaluation
  • AI governance advisory
  • EU AI Act compliance

Where Dexai is headquartered

Location

Headquarters

HQ city
Rome
HQ country
Italy
HQ region
Europe

Offices1 record

Markets served

Dexai business model

Business model
GTM type
B2B
Offering type
Services
Cost components
Personnel, Technology or R&D, Marketing or Sales, Operations

Revenue model

  1. Professional Consulting Services: Bespoke, tailored engagements across four service lines: AI Models Evaluation, AI Literacy training, AI Ethics advisory, and AI Act Compliance. Each engagement is scoped to the client's specific AI systems, business context, and regulatory risk level.
  2. EU-Funded Research Project Participation: Participation in EU-funded Horizon Europe and LIFE programme projects (ENGAGE4ESG, SOLARIS, LUCIA, AI4HOPE, CERTAIN, SHIELD) providing co-funding and project-based revenue for ethics, governance, and compliance work-packages.
  3. Training & Workshop Delivery: Delivery of AI literacy workshops, executive briefings, ethics-by-design workshops, and use-case walkthroughs as part of the AI Literacy and AI Ethics service lines.

Pricing tiers

ModelBillingPrice
OtherMulti-year contractBespoke consulting engagements

Go-to-market motion4 records

Distribution channels3 records

Marketing channels8 records

Dexai product offering

Product offering

Core offering

Dexai is an AI ethics and governance consultancy that delivers bespoke advisory engagements across four service lines: AI Models Evaluation (independent LLM benchmarking, robustness, fairness, and explainability assessments), AI Literacy (training programs tied to the EU AI Act Article 4 obligation), AI Ethics (ethics-by-design frameworks and stakeholder engagement), and AI Act Compliance (risk classification, gap analysis, conformity documentation, and audit readiness). Each engagement is scoped to the client's AI systems, business context, and regulatory exposure.

Product overview

Dexai operates as a consulting and advisory firm rather than a unified software platform, offering four core services that are typically sold as standalone engagements but commonly combined across the AI lifecycle: AI Models Evaluation (technical assessment of LLMs and other AI systems), AI Literacy (training programs tied to the EU AI Act's literacy obligation), AI Ethics (ethics-by-design and governance consulting), and AI Act Compliance (risk classification, gap analysis, and conformity support under the EU AI Act). Underpinning these services is Dexai's research initiative Icaro Lab, which investigates AI safety vulnerabilities such as poetry-based jailbreaks across state-of-the-art LLMs, and Dexai's participation in EU-funded projects (e.g., ENGAGE4ESG, SOLARIS, LUCIA, AI4HOPE, CERTAIN, SHIELD) where it contributes AI ethics, governance, and regulatory expertise.

Differentiator

Problem solved

Functional benefit

Products and services

  • AI Models Evaluation Independent, rigorous evaluation of AI models (especially large language models) covering performance benchmarking, robustness testing under perturbations and adversarial inputs, fairness and bias analysis, explainability/transparency checks, and qualitative and quantitative risk assessment aligned with the client's regulatory context.
  • AI Literacy Tailored training and education programs that build foundational AI awareness, critical thinking, and practical understanding across executives, non-technical staff, and policy/legal teams, delivered via custom learning paths, workshops, executive briefings, use-case walkthroughs, and ongoing learning support.
  • AI Ethics Consulting support for organizations to integrate ethical principles (transparency, fairness, accountability, human oversight) into the design, development, and deployment of AI systems, including ethics assessments, internal framework development, ethics-by-design workshops, stakeholder engagement guidance, and ongoing advisory support for ethics boards and review committees.
  • AI Act Compliance Strategic and operational guidance to help organizations comply with the EU AI Act, including risk classification of AI systems, gap analysis against regulatory requirements, tailored compliance roadmap, technical documentation support for conformity and audit readiness, and ongoing advisory on regulatory developments.
  • Icaro Lab Dexai's research initiative on AI safety and trustworthiness. Icaro Lab studies how poetic language can bypass AI safety guardrails in large language models and exposes blind spots in current AI safety testing methodologies, contributing to the development of more robust evaluation and red-teaming practices.

Quantifiable outcome

  • 62% attack success rate identified using handcrafted poetic prompts across 25 AI models
  • +2 more outcomes

Companies that use Dexai

Customer profile

Segments6 records

Ideal customer profiles4 records

Dexai technology and API

Technology

Technology focussed Yes

API detail

Has API
No
API docs
API detail

Core technology

AI maturity

App detail

AI capability7 records

Feature7 records

Dexai partnerships and signals

Strategic signal

Partnerships

19 partnerships are on record, tiered flagship / core research partnership, core research partner, core eu project, supporting partner via co-founder affiliation, minor — logo partner and supporting partner.

  • Icaro Labflagship / core research partnershipStrategic or Co-development PartnerIcaro Lab is a research collaboration between Sapienza University of Rome, DexAI, and Sant'Anna School of Advanced Studies. It produced the high-profile poetic jailbreak study that achieved 62% attack success rate across 25 AI models.
  • Sapienza University of Romecore research partnerStrategic or Co-development PartnerPartner in the Icaro Lab collaboration. Sapienza is also displayed on the Dexai website partner logo wall. The university co-develops research with Dexai on AI safety and ethics.
  • Sant'Anna School of Advanced Studiescore research partnerStrategic or Co-development PartnerPartner in the Icaro Lab collaboration with Sapienza University of Rome and DexAI. The collaboration produced research on bypassing AI safety guardrails via poetic language.
  • ENGAGE4ESG (LIFE programme)core eu projectStrategic or Co-development PartnerLIFE programme project aligning real estate lending with EU climate goals and ESG standards. Dexai supports development of compliant, ethical frameworks for data use and sustainability reporting.
  • SOLARIS (Horizon Europe)core eu projectStrategic or Co-development PartnerHorizon Europe project combatting disinformation and echo chambers using generative AI. Dexai contributes to the ethical design and stakeholder engagement strategy.
  • LUCIA (Horizon Europe)core eu projectStrategic or Co-development PartnerHorizon Europe project aimed at reducing cancer inequalities across Europe by developing AI-driven solutions for early detection and prevention. Dexai focuses on AI governance, ethics, and stakeholder coordination.
  • AI4HOPEcore eu projectStrategic or Co-development PartnerProject exploring AI-driven support tools for palliative care in dementia. Dexai leads activities related to legal compliance, auditing practices, and inclusive AI system design in sensitive healthcare environments.
  • CERTAIN (Horizon Europe)core eu projectStrategic or Co-development PartnerHorizon Europe initiative working on transparency and certification of trustworthy AI systems. Dexai helps define ethical and regulatory requirements and supports creation of practical tools for responsible AI lifecycle management.
  • SHIELD (Horizon Europe)core eu projectStrategic or Co-development PartnerInnovation Action under Horizon Europe (2025–2029), SHIELD focuses on early detection of pancreatic cancer in individuals with genetic predispositions, validating a new blood-based diagnostic test. Dexai contributes to ethical, legal, and social framework of the study and supports patient engagement and clinical translation.
  • PANETTA Law Firmsupporting partner via co-founder affiliationImplementation/ SI/ Consulting PartnerCo-founder Federico Sartore is a Partner at PANETTA Law Firm, advising companies on data protection, cybersecurity, and AI governance. The firm supports Dexai's legal and ethical advisory work.
  • ANSAminor — logo partnerOthersListed on Dexai's 'Our partners' logo wall on the homepage. Specific collaboration details not disclosed in source.
  • Mariboruminor — logo partnerOthersListed on Dexai's 'Our partners' logo wall on the homepage. Likely the University of Maribor or related institution. Specific collaboration details not disclosed.
  • Amsterdamminor — logo partnerOthersListed on Dexai's 'Our partners' logo wall on the homepage. Likely the University of Amsterdam or Amsterdam-based institution. Specific collaboration details not disclosed.
  • LUMSAminor — logo partnerOthersLibera Università Maria Ss. Assunta (LUMSA) — listed on Dexai's 'Our partners' logo wall. Specific collaboration details not disclosed.
  • Vicomtechsupporting partnerStrategic or Co-development PartnerSpanish applied research center listed on Dexai's partner logo wall. Likely a research consortium partner in EU projects.
  • RED (red_r)minor — logo partnerOthersListed on Dexai's 'Our partners' logo wall. Specific collaboration details not disclosed in source.
  • Digital Planetminor — logo partnerOthersListed on Dexai's 'Our partners' logo wall. Specific collaboration details not disclosed in source.
  • Humanitassupporting partnerStrategic or Co-development PartnerListed on Dexai's partner logo wall — likely Humanitas Research Hospital or Humanitas University, an Italian healthcare partner for healthcare AI projects (e.g., AI4HOPE).
  • CINI (Consorzio Interuniversitario Nazionale per l'Informatica)supporting partnerStrategic or Co-development PartnerItalian national inter-university consortium for informatics listed on Dexai's partner logo wall — likely a research and standardization partner.

Scale indicators8 records

Recent moves6 records

Expansion highlights5 records

Dexai competitors and assessment

Company assessment

Direct peers

  • Trustible: Trustible is an AI governance and policy management platform that helps enterprises operationalize responsible AI programs and map to regulations including the EU AI Act. Directly comparable buyer, problem space, and EU regulatory framing as Dexai.
  • Credo AI: Credo AI provides an AI governance platform covering AI risk management, regulatory mapping (including EU AI Act and NIST AI RMF), and audit-ready documentation. Targets the same enterprise governance, risk, and compliance buyer as Dexai and competes for AI Act compliance budgets; differentiates via SaaS platform vs. Dexai's bespoke services.
  • Fairly AI: Fairly AI provides AI risk and compliance management software with EU AI Act assessments, model evaluation, and governance workflows. Comparable to Dexai's AI Model Evaluation and AI Act Compliance services but delivered via a SaaS platform.
  • Saidot: Saidot is a Finnish AI governance SaaS focused on EU AI Act transparency, AI literacy, and stakeholder engagement tooling. Directly comparable to Dexai on AI Act compliance and AI literacy, with similar European policy-shaping positioning; differs in being more productized.
  • Holistic AI: Holistic AI is an AI governance, risk, and compliance platform that explicitly helps organizations operationalize the EU AI Act — offering AI risk assessment, bias audits, and regulatory readiness tooling. Closely comparable to Dexai in serving the same enterprise buyer navigating AI Act conformity, though Holistic AI is more product-led whereas Dexai is consulting-led.
  • Monitaur: Monitaur is an AI governance, risk, and compliance platform with model monitoring, bias testing, and audit-ready documentation. Overlaps with Dexai's model evaluation and governance framework services, especially for regulated US enterprises; serves a similar GRC buyer.

Emerging players

  • Fiddler AI: Fiddler AI provides an AI observability and explainability platform with bias monitoring and model performance tracking. Adjacent to Dexai's model evaluation and explainability/transparency work, but more productized and focused on continuous monitoring rather than conformity assessments.
  • Enzai: Enzai offers an AI governance and risk management platform focused on helping organizations comply with the EU AI Act and emerging AI regulations. Closely comparable mission and EU regulatory framing as Dexai; smaller scale and more product-focused.
  • Responsible AI Institute: Responsible AI Institute is a non-profit developing AI assurance and certification frameworks, including assessment against the EU AI Act. Comparable to Dexai's standards and conformity work, with broader certification ecosystem scope rather than consulting delivery.

Broad incumbents

  • Deloitte AI & Data Risk Services: Deloitte's Trustworthy AI practice offers enterprise-scale AI governance, EU AI Act readiness, model risk management, and responsible AI advisory. Represents the broad-incumbent competitor with deep enterprise relationships, large bench, and tool-enabled delivery that Dexai ultimately competes against for AI Act advisory budgets.

Market position

Strengths4 records

Weaknesses4 records

Competitive moat4 records

Key risks6 records

Key highlights6 records

Customer concentration

Dexai social profiles

Digital presence

Dexai compliance and trust

Trust signal

Compliance2 records

Dexai financial estimates

Financial estimate

Revenue estimate

Valuation estimate

Dexai leadership team

Management profile

Number of profiles

Profiles11 records

Dexai funding detail

Funding detail

Funding overview

Funding rounds

Investors

Funding detail is available on the Subscription and Enterprise plan.Contact sales →

Dexai M&A and investment

M&A and investment

M&A

Investments

M&A and investment is available on the Subscription and Enterprise plan.Contact sales →

Frequently asked questions about Dexai

What does Dexai do?

Dexai is an AI ethics and governance consultancy that delivers bespoke advisory engagements across four service lines: AI Models Evaluation (independent LLM benchmarking, robustness, fairness, and explainability assessments), AI Literacy (training programs tied to the EU AI Act Article 4 obligation), AI Ethics (ethics-by-design frameworks and stakeholder engagement), and AI Act Compliance (risk classification, gap analysis, conformity documentation, and audit readiness). Each engagement is scoped to the client's AI systems, business context, and regulatory exposure.

Is Dexai a public or private company?

Dexai is a private company. It is classified as founder individual operated bootstrapped and is currently operating.

When was Dexai founded?

Dexai was founded in 2021. It employs 1 to 10 people.

Where is Dexai based?

Dexai is headquartered in Rome, Italy, in the Europe region.

How does Dexai make money?

Three revenue lines are on record. Professional Consulting Services are the primary driver. The others are EU-Funded Research Project Participation and training & Workshop Delivery.

Who are Dexai's main competitors?

Direct peers on record are Trustible, Credo AI, Fairly AI, Saidot, Holistic AI and Monitaur. Emerging players are Fiddler AI, Enzai and Responsible AI Institute. Deloitte AI & Data Risk Services is listed as a broad incumbent.

Does Dexai have an API?

No public API is recorded for Dexai.

What industry is Dexai in?

Dexai's product category is AI Governance & Compliance Consulting. Its primary akta.pro industry code is BPAEAHAJ, Responsible AI, AI Governance & Compliance Services, with a secondary code of HDAAAMAE, Regulatory Readiness & Audit Automation (e.g., EU AI Act, NIST AI RMF, ISO/IEC 42001). Its NAICS code is 541720 and its SIC code is 8700.

Unlock the full company data

50 free credits on sign-up, no credit card required.

Contact sales
Live signals
ZME ScienceHow a simple poem can trick AI models into building a bombItalian researchers at Icaro Lab (a collaboration between Sapienza University of Rome, DexAI, and Sant'Anna School of Advanced Studies) have discovered that poetic language can bypass AI safety guardrails, achieving a 62% attack success rate with handcrafted poems and 43% with auto-generated ones across 25 state-of-the-art AI models including ChatGPT, Claude, and Gemini. The study reveals that AI safety filters, which rely on pattern recognition for keywords like "bomb" or "malware," fail to detect harmful content embedded in poetic metaphors and figurative language, allowing dangerous instructions to slip through. The researchers warn this finding exposes a blind spot in current AI safety testing, as existing evaluation standards assume models remain stable under minor stylistic variations when they do not.The GuardianAI’s safety features can be circumvented with poetry, research findsResearchers at Icaro Lab, an initiative from ethical AI company DexAI, found that poetry can be used to bypass AI safety guardrails in an experiment testing 20 poems containing harmful requests on 25 AI models across nine companies. The study found that 62% of poetic prompts resulted in harmful content responses, with Google's Gemini 2.5 Pro responding to 100% of such prompts while OpenAI's GPT-5 nano resisted all attempts. The vulnerability stems from how LLMs predict probable next words, as poems' non-obvious structure makes it harder to detect harmful intent, making this jailbreak method accessible to anyone rather than just sophisticated actors.FuturismScientists Discover “Universal” Jailbreak for Nearly Every AI, and the Way It Works Will Hurt Your BrainResearchers from the AI safety group DEXAI and Sapienza University of Rome discovered that AI chatbots can be easily tricked into providing harmful information using a technique called "adversarial poetry." Their study revealed that stylistic variations in poetry can bypass safety mechanisms, with certain models being successfully misled up to 90 percent of the time. The findings highlight crucial limitations in current AI safety protocols and the effectiveness of various AI models in resisting such manipulation.