BETA · GAMMA-Ray is published for review. The rules, the mapping and every rating here are experimental and will change. How a rating is made · what exists · send a correction
GAMMA-Rayv0.0.155 · beta

Visor.ai

GAMMA-Ray report · Spectrum 0.2 beta · Public tier · read 2026-11-10 · not reviewed by the company yet

Visor.ai sells a platform on which organisations build and run AI agents for customer service by voice and chat, plus an agent that scores call recordings.

1solution carded
of 4 found
11sources captured
and hashed
66provisions lit
in 8 of 8 instruments
5examples of the agent
at work, with sources

Built only from Visor.ai's own public pages and, for company basics only, the news sources named beside each fact. Visor.ai has not seen this report. Nothing was tested and no credentials were used. If you are Visor.ai and something here is wrong or missing, the correction will be made and dated, and you can claim the card to add evidence.

The agentic solutions

One card per solution. Each level is computed from the evidence by a published rule; a dimension with nothing public reads "not yet evidenced", which is not a no. A card rates a product as documented, not anyone's deployment of it.

Visor.ai AI agents (voice and chat)

AI agents that answer customer phone calls, built as flows on the platform, with test suites, live supervision and EU AI Act checks.

Visor.ai AI agentsa product, as documented
SPECTRUM 0.2 BETA · Public tier · 2026-11-10
A3AutonomyActs
π Moderate · S1·R5·A1 ×3
R2ReachIntegrated
π Moderate · S1·R5·A1 ×3
C1ControlStatedwith options on · as shipped not yet evidenced
π Moderate · S1·R5·A1 ×3
O2OpennessDocumentedwith options on · as shipped not yet evidenced
π Moderate · S1·R5·A1 ×3
DraftStage: Building (8 of 16)In production

Standout. Autonomy 3 (Acts): "those requests now run end to end as automated workflows"

Next unlock. Openness 3 (Measured): public evidence that test or evaluation results are published would get there.

Trust. FACT-Ray: 18 facts: 18 π Moderate · found by an agent, captured and hashed, checked by 2 other agents (same model family, shared repository); no human has signed it.

Standards constellation. 66 provisions lit across 8 of 8 instruments.

SalesforceEAAMMOWASPMCPAAGMMMicrosoftNISTAI Act

Evidence: 7 sources, captured and hashed · the card's page has the vault's read key · Not reviewed by its vendor yet · correct or claim this card

The full card, every answer and its evidence · the governance layer

Chat AI Agent

AI agents on WhatsApp, websites and messaging channels that answer from a synced knowledge base and hand over to live chat. No card yet (generally available): this solution has not been assessed.

Nexa

The platform's built-in agent that helps build, test and review agents, and proposes fixes that a person accepts. No card yet (live (announced as new on the home page)): this solution has not been assessed.

Quality & Insights AI Agent

Scores call recordings against a customer's own scorecard and turns the scores into coaching and contact-reason reports. No card yet (generally available): this solution has not been assessed.

Our reading

The shape. Visor.ai's AI agents read A3 · R2 · C1 · O2 (Building). A customer builds each agent as a flow on a visual canvas, with the model and speech services set on each node. The voice agent connects to telephony and to the customer's contact-centre platform; the chat agent answers from a knowledge base built from the customer's own documents and kept in sync with them. Nexa, the platform's own agent, reads real conversations on a schedule and proposes fixes.

Where the evidence is strongest. A named customer shows the agent acting. At Generali Tranquilidade, common requests such as duplicate documents, address changes, adding a vehicle and cancellations run end to end as automated workflows through web services into the insurer's systems, and an advisor picks up in live chat when a person is needed. Visor.ai's customer stories give the figures: 80,000 broker conversations a month with the agent at Generali, and Max at Millennium BCP starting at 65% automation in 2021, with a goal of 80%. They are the customers' figures as Visor.ai publishes them.

What holds as shipped, and what with options on. Acts and Integrated rest on the product pages and the Generali deployment. Control 1 holds only with the Issue Agent on: the customer writes in plain language what the chat agent must not handle, and matching conversations are flagged for a person. Openness 2 rests on a reliability note saying the models can occasionally fail, the knowledge base, and the security page's list of AI providers (Azure, Vertex AI Gemini, Deepgram, AssemblyAI, ElevenLabs). Its first step, a written statement of what the agents are for, is stated for healthcare deployments, so Openness is marked available too; a purpose stated for every deployment would make it hold as shipped.

What would move it. Autonomy 4: a page saying the agent works on its own and a person approves only critical decisions. Reach 3: a published agent interface, such as an MCP server or an agent card, with documented authorisation narrower than the whole account. Control 2: a page saying which high-impact actions of the customer-facing agent need a person's approval, with a list of what it can reach. Openness 3: published evaluation results; the llms.txt at the site root already gives the machine-readable half of the rule.

Seen, and not counted. The challenger lowered several credits a reader may meet on the site. Nexa's approval of the changes it proposes governs edits to an agent, not the actions the customer-facing agent takes. Take moves one live call to a person, which is escalation, not a way to pause or stop the agent. Watchtower monitoring and Nexa's alerts are shown with sample data, not in operation, and the Super Agent routes requests to specialised agents rather than coordinating them. Headline figures such as an automation rate of up to 85% are listed as announced.

What would move it

The one piece of public evidence that would lift each card next:

The agent at work

Examples from public pages that name a customer or give figures. Each source was captured and hashed on the date shown.

Generali Tranquilidade

Chat AI agent for customers and brokers; common requests (duplicate documents, address changes, adding a vehicle, cancellations) run end to end through web services; 80,000 broker conversations a month. "conversations between brokers and the AI agent every month"

Source (captured 2026-11-10T18:45Z, sha256 5ad19f4d69ad64f9)

Generali Tranquilidade

Requests run as automated workflows connected to the insurer's systems. "those requests now run end to end as automated workflows"

Source (captured 2026-11-10T18:45Z, sha256 5ad19f4d69ad64f9)

Millennium BCP

Max, a Chat AI Agent for the bank's sales network of more than 5,000 people, with live-chat handover to advisors. "Max, an AI agent built on Visor"

Source (captured 2026-11-10T18:45Z, sha256 bf836623286299e0)

Millennium BCP

Automation of Max's questions started at 65% in 2021, with a goal of 80%. "Max started at 65% automation in 2021, and the goal is 80%"

Source (captured 2026-11-10T18:45Z, sha256 bf836623286299e0)

Farmácia Sá da Bandeira

Home page figure of 77% interaction success rate attributed to this customer. "Interaction Success Rate"

Source (captured 2026-11-10T10:41Z, sha256 03497c0c550e4bc7)

The company

The people

Public and professional only, from the company's own pages or a public professional source.

The graph

The company, its people and the stories on the left; each solution, the captured sources behind its card, the indicators those sources evidence and the provisions of the instruments those indicators touch, to the right. "Touched" never means "complies".

people and storiescompanysolutions7 sources18 indicators66 provisions, by instrumentI03: Generali's own team builds workflows and automated services on the platform without a developer, connected to the insurer's systems through web services.I08: At Generali, the platform was connected to the insurer's systems through web services, so four common requests (duplicate documents, address changes, adding a vehicle, cancellations) run end to end as automated workflows.I10: The voice agent connects to telephony and to the customer's contact-centre or CCaaS platform.I11: Generali connected the platform to its own systems through web services.I17: At Generali the chat agent handles conversations from start to finish; when a conversation needs a person, an advisor picks it up in live chat.I20: For healthcare, the agents' scope is administrative contact, written down before work starts.I22: A reliability note states that the models can occasionally fail and that providers do not guarantee total availability.I36: Each agent runs twelve automatic checks against the EU AI Act and the GDPR, each citing its article.I37: In Visor's deployment, the agent is tested end to end and validated by the customer's team in real conditions before launch.I40: A chat that needs a person moves to live chat and is routed to the right department with its history.I47: For healthcare, the page states as a rule that the agents do not give clinical advice and are set up to pass a clinical question to a person.I58: The chat agent answers from a knowledge base built from the customer's documents and kept in sync with them.I61: The security page lists the AI providers in the voice and chat path (Azure, Vertex AI Gemini, Deepgram, AssemblyAI, ElevenLabs) and what each processes.I66: In Visor's deployment, Visor trains the customer's people to run the platform.I67: The Visor success story states that Max started at 65% automation in 2021 and that the goal is 80% of questions and processes resolved without a person; only the 65% is a measured figure, and the page markup attributes the story to an archived copy of 2022-08-20.I68: Nexa reads real conversations on a schedule, spots patterns and proposes fixes.S1: Generali Tranquilidade is named as a customer; the page gives 80,000 conversations a month between brokers and the AI agent, an undated figure that the page markup attributes to an archived copy of 2022-08-20.S3: An llms.txt file resolves at the site root and indexes the product pages with markdown twins.Visor.aiGonçalo ConsiglieriBruno MatiasAfonso FonsecaRita AnjoGenerali TranquilidadeGenerali TranquilidadeMillennium BCPMillennium BCPFarmácia Sá da BandeiraVisor.ai AI agents (voice and chat)Voice and chat agentsA3 · R2 · C1 · O2Chat AI AgentChat AI AgentNexaNexaQuality & Insights AI AgentQuality & Insights AI Agentvisor.ai/success-stories/generalivisor.ai/voice-agentvisor.ai/healthcarevisor.ai/chat-agentvisor.ai/securityvisor.ai/success-stories/millenni…visor.ai/llms.txtIt is configured on a low-code or no-code builder and acts on real organisational dataI03It executes actions itselfI08It works across several domains or systemsI10It connects to external tools or services across a trust boundaryI11The agent handles routine tasks; the person is consulted on exceptionsI17Its purpose and permitted tasks are stated in writingI20Its limits and known failure cases are documented for those who use itI22The laws that apply to it are identifiedI36New agents pass an approval or review before they runI37There is a defined escalation pathI40A limit on it is a rule in proseI47The data it uses is describedI58The models and third-party components it uses are listedI61The people who work with it are trainedI66Its value is measuredI67It is improved on a regular cycleI68It is in production: a public source names a customer or gives usage figures for what it does on its ownS1It is machine-readable for agents: it publishes an llms.txt, an AGENTS.md, an API description or an agent cardS3Salesforce 13EAAMM 4OWASP 7MCP authorization 2AAGMM 3Microsoft 18NIST AI RMF 17EU AI Act 2OWASP, AT3 Citizen Developer AgentSalesforce, Level 2: Simple Orchestration, Single DomainSalesforce, Level 1, how to advance from Level 1 to Level 2Microsoft, 200: Repeatable, Business strategySalesforce, Level 3: Complex Orchestration, Multiple DomainSalesforce, Level 2, how to advance from Level 2 to Level 3Salesforce, Level 3, how to advance from Level 3 to Level 4Microsoft, 400: Capable, Business strategyMicrosoft, 200: Repeatable, Technology and dataOWASP, AT6 Externally Extended AgentNIST AI RMF, GOVERN 6.1NIST AI RMF, MAP 4.1EAAMM, L3 Consultative (survey item Q9)NIST AI RMF, MAP 1.1NIST AI RMF, MAP 2.1NIST AI RMF, MAP 3.3NIST AI RMF, MAP 1.3EU AI Act, Article 13NIST AI RMF, MAP 2.2NIST AI RMF, MEASURE 2.5NIST AI RMF, GOVERN 1.1OWASP, Maturity Level 4 Adaptive, Self-Regulating Governance, key enterprise actionMicrosoft, 100: Initial, AI governance and securityOWASP, Maturity Level 1 Experimentation without Guardrails, key enterprise actionMicrosoft, 400: Capable, AI governance and securityNIST AI RMF, MANAGE 1.1EAAMM, Prerequisite, L2 to L3NIST AI RMF, GOVERN 3.2AAGMM, GD-06OWASP, Maturity Level 0 Unaware and Ad Hoc, key enterprise actionAAGMM, GD-05EAAMM, Readiness dimension 2Salesforce, Level 0, how companies can advance from Level 0 to Level 1Salesforce, Level 1, how to advance from Level 1 to Level 2OWASP, Maturity Level 2 Policy-Defined, Human-in-the-Loop, key enterprise actionNIST AI RMF, GOVERN 2.2EU AI Act, Article 4EAAMM, Readiness dimension 3Microsoft, 100: Initial, Organization and cultureMicrosoft, 200: Repeatable, Organization and cultureMicrosoft, 300: Defined, Technology and dataMicrosoft, 300: Defined, Organization and cultureNIST AI RMF, MAP 3.4Salesforce, Level 0, how companies can advance from Level 0 to Level 1Salesforce, Level 0, how companies can advance from Level 0 to Level 1Salesforce, Level 1, how to advance from Level 1 to Level 2Salesforce, Level 3, how to advance from Level 3 to Level 4Salesforce, Level 4, how to maximize valueSalesforce, Level 4, how to maximize valueMicrosoft, 200: Repeatable, Business strategyMicrosoft, 300: Defined, AI strategy and experienceMicrosoft, 300: Defined, Business strategyMicrosoft, 400: Capable, Business strategyMicrosoft, 400: Capable, AI strategy and experienceNIST AI RMF, MAP 1.4NIST AI RMF, MAP 3.1Microsoft, 100: Initial, Organization and cultureAAGMM, GD-12Microsoft, 400: Capable, AI governance and securityMicrosoft, 500: Efficient, Business strategyMicrosoft, 500: Efficient, AI strategy and experienceNIST AI RMF, MANAGE 4.2OWASP, Maturity Level 1 Experimentation without Guardrails, key enterprise actionNIST AI RMF, MEASURE 4.3MCP authorization, Authorization (draft)MCP authorization, Authorization (draft)

The evidence

SourceReadSHA-256 (first 16)
www.visor.ai/2026-11-10T10:41Z03497c0c550e4bc7
www.visor.ai/chat-agent/2026-11-10T18:45Z25d94e3335f5dc6e
www.visor.ai/company/2026-11-10T10:41Zf4592014a8d289e8
www.visor.ai/healthcare/2026-11-10T18:45Zbea305021fe6b3ab
www.visor.ai/llms.txt2026-11-10T18:45Z35a955f100a9cf45
www.visor.ai/privacy-policy/2026-11-10T10:41Z0055c5fcebc30d78
www.visor.ai/pt/prr/2026-11-10T10:41Z52a248dec2398acc
www.visor.ai/security/2026-11-10T18:45Zf337c67022110924
www.visor.ai/success-stories/generali/2026-11-10T18:45Z5ad19f4d69ad64f9
www.visor.ai/success-stories/millennium-bcp/2026-11-10T18:45Zbf836623286299e0
www.visor.ai/voice-agent/2026-11-10T18:45Zc641b9f6a84c7399

How this report was made

A researcher-assessor gathered and captured the sources and mapped the evidence to indicators, without assigning a level. A challenger argued against every level three and four and every doubtful credit. The deterministic scorer computed the levels and the gate checked the rules. A card appears here only after two rounds of review and a sign-off by agents that did not write it (the team's workflow); Visor.ai AI agents (voice and chat): reviewed in 4 rounds, signed off by gamma.approve@web-summit.dinis on 2026-11-10. A person's sign-off of the series is still to come.

RoleWho
researcher-assessora sub-agent of gamma.studio@web-summit.dinis
challengera sub-agent
scorerbuild/spectrum.mjs
analyst, designer, writergamma.studio@web-summit.dinis
reviewgamma.review@web-summit.dinis, 4 rounds
sign-offgamma.approve@web-summit.dinis, 2026-11-10

The vault

Every capture, the dossier, both assessors' mappings, the challenger's notes and this report's data are in a public sgit vault (w4byv6dr). Read key:

sgit_public_read_90a76763f617c725a429b7b445125e43a01bf7f880c08904b7803018bc65935b:w4byv6dr

With sgit: sgit clone followed by the key. The key reads; it cannot write.

Corrections

None yet. A correction is made in place, dated here, and carried to the card, the vault and the article.

The article

Visor.ai agents on the GAMMA-Ray Spectrum: A3 · R2 · C1 · O2, written from this report for PrompTugal.news.