WitnessAI announced on January 13, 2026, that it raised $58 million in strategic funding led by Sound Ventures. The announcement also introduced expanded controls for AI agents, Model Context Protocol (MCP) servers and tool calls. The company says the financing will fund global go-to-market expansion and product development.
The round is notable because WitnessAI is targeting a newer security problem: systems that can retrieve data, call software tools and execute workflows, not merely answer chatbot prompts.
What WitnessAI announced
The financing was described as a “strategic funding” round, not as a Series B or another conventional venture designation. Sound Ventures led the round. The official announcement names Fin Capital, Qualcomm Ventures, Samsung Ventures and Forgepoint Capital Partners as participants.
| Item | Disclosed detail |
|---|---|
| Announcement date | January 13, 2026 |
| Amount | $58 million |
| Lead investor | Sound Ventures |
| Other named investors | Fin Capital, Qualcomm Ventures, Samsung Ventures and Forgepoint Capital Partners |
| Stated use of proceeds | Global go-to-market expansion and product development |
| Prior disclosed round | $27.5 million Series A announced in May 2024, co-led by GV and Ballistic Ventures |
| Disclosed total from those two rounds | $85.5 million; this does not establish the company’s complete lifetime financing |
WitnessAI’s announcement does not disclose valuation, dilution, financing structure, dollar revenue, customer retention, gross margin or the number of paying customers. The company reported more than 500% year-over-year ARR growth, fivefold headcount growth and protection for hundreds of thousands of enterprise employees and apps; those figures are company-reported and unaudited.
Do these 3 things before closing this tab:
1Repair Windows errors before they cause bigger problems2Scan for outdated or missing drivers - takes under a minute3Clear out junk files and repair common Windows errors#1 Best Overall
- POWERFUL SECURITY KEY: The Security Key C NFC is the essential physical passkey for protecting your digital life from phishing attacks. It ensures only you can access your accounts.
- WORKS WITH 1000+ ACCOUNTS: Compatible with Google, Microsoft, and Apple. A single Security Key C NFC secures 100 of your favorite accounts, including email, password managers, and more.
- FAST & CONVENIENT LOGIN: Plug in your Security Key C NFC via USB-C and tap it, or tap it against your phone (NFC) to authenticate. No batteries, no internet connection, and no extra fees required.
- TRUSTED PASSKEY TECHNOLOGY: Uses the latest passkey standards (FIDO2/WebAuthn & FIDO U2F) but does not support One-Time Passwords. For complex needs, check out the YubiKey 5 Series.
- BUILT TO LAST: Made from tough, waterproof, and crush-resistant materials. Manufactured in Sweden and programmed in the USA with the highest security standards.
What WitnessAI sells
WitnessAI describes its product as a unified platform for observing, controlling and protecting enterprise use of AI. Its coverage is intended to span employee-facing applications, internal and customer-facing AI systems, models, autonomous agents, MCP servers and external tools.
Observe
The company says it discovers AI applications, agents and MCP servers, then records interactions and activity across network-level traffic. Its positioning includes native applications and workflows outside a browser-only environment. Details are described on the Observe page.
Control
WitnessAI says administrators can enforce policies, route requests according to risk, cost or purpose, redact or tokenize sensitive data, govern tool access and create audit trails. The company emphasizes behavior and intent-based classification rather than relying only on keywords or regular expressions. That is a product-positioning claim, not independent proof of superior detection.
Rank #2
- POWERFUL SECURITY KEY: The YubiKey 5C NFC is the most versatile physical passkey, protecting your digital life from phishing attacks. It ensures only you can access your accounts
- WORKS WITH 1000+ ACCOUNTS: Compatible with popular accounts like Google, Microsoft, and Apple. A single YubiKey 5C NFC secures 100+ of your favorite accounts, including email, password managers, and more
- FAST & CONVENIENT LOGIN: Plug in your YubiKey 5C NFC via USB and tap it, or tap it against your phone (NFC), to authenticate. No batteries, no internet connection, and no extra fees required
- MOST SECURE PASSKEY: Supports FIDO2/WebAuthn, FIDO U2F, Yubico OTP, OATH-TOTP/HOTP, Smart card (PIV), and OpenPGP. That means it’s versatile, working almost anywhere you need it
- PRIMARY & SPARE KEYS: Just like having a spare house key, we recommend buying two YubiKeys - one for daily use and one as a spare. That way you’ll never get locked out of your accounts
Protect
The Protect offering is described as bidirectional runtime defense: inspect prompts before they reach a model or agent, detect prompt injection and jailbreak attempts, filter responses, protect sensitive information and support automated red teaming. “Blocks” or “protects” should be read as stated capabilities; no public evidence establishes that every attack is stopped.
The Tool Desk
Outbyte Driver Updater FREEScan for outdated or missing drivers - takes under a minuteDriver Scan →Outbyte PC Repair FREERepair Windows errors before they cause bigger problemsFix Now →What changed with the agentic-security launch
The January announcement paired the funding with expanded Agentic Security capabilities. The distinction matters because an agent can turn a natural-language instruction into an external action.
Visibility before and during execution
- Which agents are active and which human or service identity is associated with them.
- Which MCP servers and tools are available or used.
- What data an agent shares and which commands it executes.
- The surrounding decision context, rather than only an isolated prompt or response.
Runtime controls
- Prompt inspection before an agent acts.
- Rules applied at the MCP-server and individual tool-call level.
- Detection of prompt injection, including multi-turn attack patterns.
- Filtering of agent responses before users see them or downstream systems receive them.
- Real-time tokenization of sensitive information.
- Audit records linking users, agents, tools and policies.
These are separate controls. A platform that filters output is not automatically authorizing tool calls, and an inventory of agents is not the same as a pre-execution block. Buyers should test each function independently.
Rank #3
- POWERFUL SECURITY KEY: The YubiKey 5 NFC is the most versatile physical passkey, protecting your digital life from phishing attacks. It ensures only you can access your accounts
- WORKS WITH 1000+ ACCOUNTS: Compatible with popular accounts like Google, Microsoft, and Apple. A single YubiKey 5 NFC secures 100+ of your favorite accounts, including email, password managers, and more
- FAST & CONVENIENT LOGIN: Plug in your YubiKey 5 NFC via USB and tap it, or tap it against your phone (NFC), to authenticate. No batteries, no internet connection, and no extra fees required
- MOST SECURE PASSKEY: Supports FIDO2/WebAuthn, FIDO U2F, Yubico OTP, OATH-TOTP/HOTP, Smart card (PIV), and OpenPGP. That means it’s versatile, working almost anywhere you need it
- PRIMARY & SPARE KEYS: Just like having a spare house key, we recommend buying two YubiKeys - one for daily use and one as a spare. That way you’ll never get locked out of your accounts
Why agent security is a different problem
Traditional controls often monitor network traffic, files, identities or endpoint events. Tool-using AI introduces another question: what was the system trying to do, and who authorized that intent?
- Who approved the action?
- Which agent performed it, under which identity?
- What tools and data were within its reach?
- Was the instruction malicious, ambiguous or manipulated?
- Could the action be stopped before execution?
- Can investigators reconstruct the sequence afterward?
That gap is the market problem WitnessAI is pursuing. MCP and agent security are still evolving categories, so the company’s launch should be viewed as an attempt to create a control and accountability layer for semi-autonomous systems, not as a settled industry standard.
What the investor group may signal
Sound Ventures’ participation is strategically notable because WitnessAI’s release identifies the firm’s investments in companies including OpenAI, Anthropic and SentinelOne. Those prior investments do not validate WitnessAI’s technology or guarantee its future performance.
Rank #4
- POWERFUL SECURITY KEY: The Security Key NFC is the essential physical passkey for protecting your digital life from phishing attacks. It ensures only you can access your accounts.
- WORKS WITH 1000+ ACCOUNTS: Compatible with Google, Microsoft, and Apple. A single Security Key NFC secures 100 of your favorite accounts, including email, password managers, and more.
- FAST & CONVENIENT LOGIN: Plug in your Security Key NFC via USB-A and tap it, or tap it against your phone (NFC) to authenticate. No batteries, no internet connection, and no extra fees required.
- TRUSTED PASSKEY TECHNOLOGY: Uses the latest passkey standards (FIDO2/WebAuthn & FIDO U2F) but does not support One-Time Passwords. For complex needs, check out the YubiKey 5 Series.
- BUILT TO LAST: Made from tough, waterproof, and crush-resistant materials. Manufactured in Sweden and programmed in the USA with the highest security standards.
Qualcomm Ventures and Samsung Ventures highlighted the need to secure AI as models move across cloud and edge environments. Fin Capital and SMBC emphasized accountability and financial-services use cases. Those comments suggest interest across device, cloud, financial and cybersecurity markets, but they are not evidence of customer adoption.
Does WitnessAI replace conventional security tools?
WitnessAI presents a unified platform as an alternative to stitching together a secure web gateway, DLP, identity and access management, API gateway, SIEM/SOAR, endpoint protection, model guardrails and separate red-team tooling. A single policy and audit layer could reduce fragmented attribution, but it also creates vendor concentration and a larger impact if the platform is misconfigured or unavailable.
Network-level coverage may avoid application rewrites, yet buyers should verify what context remains available through encrypted traffic, streaming responses, custom agent frameworks and proprietary tool calls. Intent-based rules may be more flexible than keyword matching, but administrators need explainable, reproducible decisions and an approval path for ambiguous actions.
Outdated Drivers Are Slowing You Down
One free scan finds every outdated or missing driver and matches the right update for your exact hardware.Free scan · exact hardware matchPC Slower Than It Used to Be?
A free scan shows the junk files, broken settings and background clutter dragging Windows down - then fixes them in one click.Free scan · Windows 10 & 11Best Value
- The information below is per-pack only
- POWERFUL SECURITY KEY: The Security Key C NFC is the essential physical passkey for protecting your digital life from phishing attacks. It ensures only you can access your accounts.
- WORKS WITH 1000+ ACCOUNTS: Compatible with Google, Microsoft, and Apple. A single Security Key C NFC secures 100 of your favorite accounts, including email, password managers, and more.
- FAST & CONVENIENT LOGIN: Plug in your Security Key C NFC via USB-C and tap it, or tap it against your phone (NFC) to authenticate. No batteries, no internet connection, and no extra fees required.
- TRUSTED PASSKEY TECHNOLOGY: Uses the latest passkey standards (FIDO2/WebAuthn & FIDO U2F) but does not support One-Time Passwords. For complex needs, check out the YubiKey 5 Series.
Alternatives and comparison points
| Approach | Likely strength | Questions to compare with WitnessAI |
|---|---|---|
| HiddenLayer | AI discovery, supply-chain security, model and workflow protection, runtime security and governance | Does it provide equivalent workforce discovery, MCP governance and network-level enforcement? |
| Lakera | AI-native protection for generative-AI and agent applications, experiences and user data | How do deployment, employee coverage, pre-execution tool controls and policy administration compare? |
| Existing security stack | Reuses current contracts, identities, dashboards and operating expertise | Can separate tools map one human identity across prompts, agents, tools and model actions? |
WitnessAI may complement rather than eliminate these products. A proof of concept should establish whether its broad AI view adds meaningful control beyond existing DLP, proxy, identity and application-security systems.
Public pricing and likely buyer fit
WitnessAI uses a demo-led sales process. Its AWS Marketplace listing provides list-price signals, but these are not universal negotiated quotes and may represent specific packages and minimum commitments.
| Marketplace package | Minimum commitment | Displayed 12-month price |
|---|---|---|
| WitnessAI for Enterprise | 1,000 users | $180 per user, or $180,000 annually |
| WitnessAI Agentic Visibility | 2,500 agents | $24 per agent, or $60,000 annually |
| WitnessProtect – Model Protection Guardrail | Not stated | $300,000 annually |
| Additional usage | Usage-based | $0.01 per unit; AWS infrastructure may apply separately |
The listing says pricing is subject to contract terms and is non-cancellable and non-refundable except where required by law. See the AWS Marketplace listing for the displayed terms.
Likely fit
- Large or regulated enterprises with many AI users, applications or agents.
- Security teams needing centralized policy, identity linkage and audit records.
- Organizations prepared to fund integration, testing and ongoing policy operations.
Likely poor fit
- Small teams seeking transparent monthly pricing.
- A single developer wanting a lightweight API guardrail.
- Organizations with only one model and no autonomous tool use.
Questions to answer in a proof of concept
- Can the platform see browser, native desktop, API, IDE, cloud and agent traffic, including private MCP servers?
- Can it stop a malicious or unauthorized tool call before execution, rather than only log it?
- Can every action be tied to a human, service account, application, agent and tool?
- How are prompt injection, data exfiltration and encoded or multi-turn attacks detected?
- What are the measured false-positive, false-negative and latency rates on your own traffic?
- What prompts, responses and metadata are retained, where are they processed, and who can access them?
- Does deployment require endpoint agents, certificates, proxies, SDK changes or network-routing changes?
- What happens during an enforcement outage, provider switch or policy update?
- Can logs be exported to your SIEM, SOAR, GRC and retention systems?
- Are usage overages, AWS infrastructure, implementation and professional services charged separately?
Bottom line for enterprise buyers
The $58 million round demonstrates investor interest in enterprise AI security, but funding alone does not prove technical efficacy or commercial durability. The more consequential development is WitnessAI’s effort to govern employee and agentic AI activity through one observation, enforcement and audit layer.
The buying decision turns on evidence from a controlled deployment: coverage across real applications and agents, reliable pre-execution tool controls, explainable policy decisions, privacy protections, manageable latency and a total cost that justifies replacing or integrating existing controls.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
