Cognizant joins Qualys alliance to fix vulnerabilities at AI speed
Cognizant has joined the Qualys Global mROC Alliance to run AI-enabled risk operations, as Qualys data shows only 0.74% of 2025's 48,172 disclosed vulnerabilities were actively weaponised.
Why it matters
- Cognizant joined the Qualys Global mROC Alliance on November 19, 2026.
- Microsoft's November Patch Tuesday included 1,093 vulnerabilities, up from 93 before frontier AI models emerged.
- Only 357 of 48,172 vulnerabilities disclosed in 2025 (0.74%) were confirmed remotely exploitable and actively weaponised.
- Qualys said its system delivered 90% of detections within minutes of disclosure during November's Patch Tuesday.
- More than 10,000 organisations use the Qualys platform, including most of the Fortune 100.
Microsoft's November 2026 Patch Tuesday carried 1,093 vulnerabilities in a single release — up from 93 before frontier AI models emerged — and that surge is the backdrop for Cognizant joining the Qualys Global mROC Alliance, announced November 19, 2026.
The partnership expands an existing strategic relationship into managed cyber risk operations. It combines Cognizant's cyber vulnerability management services with Qualys Enterprise TruRisk Management and TruRisk Eliminate with TruConfidence. The target: help enterprises detect, prioritise and remediate cyber risk across AI, cloud and hybrid environments.
Why the gap between disclosure and threat matters
The numbers explain the urgency. Qualys said attackers using AI tools can turn a new advisory into a weaponised attack in as little as four hours. Yet the volume of disclosed flaws vastly outweighs the number that actually matter.
Qualys' Threat Research Unit found that of 48,172 vulnerabilities disclosed in 2025, only 357 — 0.74% — were confirmed as both remotely exploitable and actively weaponised.
That gap has become a central problem for security teams. They must decide what to fix first, and how fast, without disrupting business systems. Patch everything and you break production. Patch nothing and you invite breach. The alliance is built around closing that distance.
How the managed risk operations model works
Under the arrangement, Cognizant will deliver an AI-enabled risk operations centre for joint clients, built on Qualys Enterprise TruRisk Management. The service works within existing customer tools, change windows and approval processes.
Qualys' autonomous agents detect, validate and remediate exposures. Cognizant's cybersecurity teams run the programme end to end. The model moves through three stages: detection, prioritisation and remediation.
Detection. Qualys InstaScan, using Agent Insta, identifies affected assets from existing inventory data without waiting for a scan. Qualys said the system delivered 90% of detections within minutes of disclosure during November's Patch Tuesday. Cognizant's service monitors new disclosures daily and routes findings to the relevant asset owners.
Prioritisation. Qualys Agent Val assesses which vulnerabilities are exploitable in a client's specific environment. The stage includes cyber risk quantification to estimate financial impact, giving business leaders a basis for resource allocation. Cognizant maintains the validated list and reviews remediation priorities with clients weekly.
Remediation. TruRisk Eliminate with TruConfidence accounts for existing security controls before selecting a fix — patch, mitigation or isolation. Qualys said the system verifies whether the fix worked and can roll back if system health declines. Cognizant agrees scope and approvals with client application owners, and phases in automation starting from lower-risk systems.
What Cognizant will actually sell
Cognizant plans to package the work into four managed risk operations services built on Qualys Enterprise TruRisk Management:
- Cyber risk quantification and advisory
- Onboarding and integration
- Managed risk monitoring
- Managed risk remediation
The structure signals a wider shift in cybersecurity services toward managed operating models designed to narrow the gap between finding vulnerabilities and fixing them. Large organisations typically run multiple security tools and formal change management processes, which slows remediation even when a threat is urgent.
For service providers, that friction creates an opening to sell tightly integrated workflows that combine risk scoring, operational oversight and approved remediation steps. It also raises the stakes on a harder question: whether automated decision-making can work safely in production environments.
What the executives said
Vishal Salvi, senior vice president and global head of cybersecurity at Cognizant, framed the alliance around the weakness of piecemeal controls.
"Our 2026 Cybersecurity Pulse Poll shows that strength in one control, such as detection or remediation, still leaves enterprises exposed in the AI era," Salvi said. "Together with Qualys, we are connecting detection, prioritisation and remediation into one managed operating model, designed to help clients manage risk, strengthen resilience and support the adoption of AI with greater confidence."
Qualys positioned the alliance as part of its push to expand managed risk operations through partners rather than only through direct software sales. The company said more than 10,000 organisations use its platform globally, including most of the Fortune 100.
Shailesh Athalye, chief product solutions officer at Qualys, said: "Qualys Enterprise TruRisk Management and TruRisk Eliminate help enterprises in the frontier AI era detect risk at the speed of disclosure, validate which exposures are real and remediate autonomously in a way customers can trust."
"Cognizant brings deep process and engineering expertise in vulnerability management and remediation, balancing speed with resiliency," Athalye said. "Together, we operationalise the AI-native Risk Operations Centre and help customers remediate at AI speed."
Can autonomy survive the change board?
The bet here is that autonomy and governance can coexist. TruRisk Eliminate can roll back a failed fix automatically, but it still operates inside client approval processes that Cognizant negotiates system by system. The service starts automation on lower-risk systems and phases upward, which is the realistic path — few enterprises will hand production change control to an agent on day one.
The 0.74% figure cuts both ways. It justifies aggressive triage, since almost nothing disclosed is actively weaponised. But identifying that 357-vulnerability sliver before attackers do, within a four-hour window, is precisely the job AI-driven prioritisation claims to do.
Whether the model holds up will show in remediation time, not detection counts. If the alliance shortens the interval between disclosure and verified fix inside change-controlled enterprises, expect more services firms to copy the structure.
Source: itbrief.com.au
More from Elena Vasquez
Show full bio
Market editor covering technology and product news in artificial intelligence and machine learning at AI In Context.
744 articles
Related articles
- IBM Plans Claude Mythos Scans Inside Concert Protect
- Cohere Buys AISLE and Cubic, Extending Sovereign AI Into Cyber Defense
- Nine in 10 Australian firms hit by slow patching risks
- 85% of Security Leaders Report Patching-Delay Incidents, Azul Survey Finds
- Google says 2027 is the year AI agents reshape enterprise security