Policy & Regulation

Knightcorp report warns tech insurers leave AI and cloud claims uncovered

Knightcorp's Forerunner Report warns that policies silent on AI, undeclared cloud systems and new device rules can turn technology insurance claims into disputes.

Knightcorp warns tech insurers over AI & cloud gaps
Knightcorp warns tech insurers over AI & cloud gapsTumisu / pixabay
By Rebecca Stone5 min read

Why it matters

  • Knightcorp launched its Forerunner Report – Technology covering AI, cyber, cloud, directors' liability and connected devices.
  • The report cites Lander & Rogers finding that AI exclusions are appearing in some policy wordings.
  • It includes an 11-question insurance readiness checklist and a broker checklist for AI risk.
  • ASIC greenwashing penalties hit Mercer Superannuation and Vanguard Investments.
  • In the Medibank matter, a Federal Court found Deloitte post-incident reports were not covered by legal professional privilege.

Australian broker Knightcorp has launched its Forerunner Report – Technology, and its central warning is blunt: some professional indemnity and cyber policies are silent on AI, and that silence can turn a claim into a dispute.

The report examines insurance issues across five connected pressure points for technology businesses: artificial intelligence, cyber risk, cloud exposure, directors' liability and connected devices. Its core argument is that insurance reviews are failing to keep pace with how technology companies now operate, particularly when those companies depend on AI tools, cloud infrastructure and third-party software platforms.

Knightcorp frames the stakes in governance terms. The report presents insurance not as an annual procurement exercise but as part of broader risk governance for finance leaders, boards and technology teams. For a sector where product stacks change faster than annual policy renewals, that framing carries direct financial consequences.

Are policies ready for AI claims?

The report's central finding concerns policy wording. Some professional indemnity and cyber policies do not clearly address AI-related claims, Knightcorp said. The gap matters in two scenarios: when businesses build AI into their own products, and when they use AI embedded in third-party platforms.

The market is already moving. Citing law firm Lander & Rogers, the report says AI exclusions are appearing in some policy wordings. That development adds pressure on companies to check whether their existing insurance arrangements match how they actually use the technology.

Trent Brown, Chief Executive Officer of Knightcorp, said the issue for companies is whether insurance arrangements still reflect the pace of change inside technology businesses.

"The question is no longer simply 'Are we insured?' It is 'Do our risk and insurance arrangements still reflect the business we have become?' In technology, a business can change faster than the systems around it," Brown said.

Brown said uncertainty is most likely to surface in policy interpretation and claims handling, particularly once an incident has already occurred. By then, the wording a company bought is the wording it must live with.

Justin Geldart, Account Director at Knightcorp, drew a distinction that many technology buyers have not yet tested against their own cover.

"Where a policy is silent on AI, the outcome is uncertain and coverage can be disputed. We encourage technology leaders to ask how their cover treats AI they build versus AI they buy, because the answer can differ," Geldart said.

What happens with cloud and vendor systems?

The report applies similar scrutiny to cloud and vendor platforms in cyber cover. Undeclared systems, or systems that do not fit a policy's wording, can become a point of dispute during claims, Knightcorp said.

The recommended starting point is practical. The report suggests businesses first map the third-party systems that handle client data, then test how each one is treated in policy applications and in the wording itself. A system the insurer never knew about is a system the insurer may not pay for.

Claims outcomes often turn on what was disclosed to insurers at the outset and how the policy defines the underlying risk, Knightcorp said. That makes the accuracy of applications and schedules a material issue for businesses with complex and fast-changing software stacks.

Jessica Tanner, Key Account Manager, Technology, at Knightcorp, put the test in operational terms.

"Claims tend to be tested on what was declared and how the policy defines it. For a technology business, the practical question is whether your wording and your application reflect the platforms your business actually relies on," Tanner said.

Do connected devices face new security rules?

The report extends the same disclosure logic to connected hardware. Companies that make or supply smart devices should check whether new mandatory security standards apply to them.

One rule the report names directly: bans on universal default passwords on in-scope consumer devices manufactured after a specified threshold. The report also warns that undeclared connected devices may affect how insurers assess a claim. A smart device sold into the market without being reflected in the insurance schedule carries the same dispute risk as an undeclared cloud platform.

What is the board-level exposure?

Governance and disclosure risks form another section of the research. Knightcorp highlights recent action by the Australian Securities and Investments Commission over misleading sustainability claims. Penalties in greenwashing cases involving Mercer Superannuation and Vanguard Investments fell on the companies involved, the report notes.

That enforcement record raises wider questions for boards about how they substantiate public governance statements. The report advises organisations to examine how directors' and officers' cover would respond if directors were drawn into proceedings, and whether that insurance sits in a standalone policy or within management liability cover.

Regulatory change compounds the pressure. The report notes that organisations covered by the Privacy Act that use automated decision-making in certain ways will need to disclose that use in their privacy policies.

A Federal Court development adds a further caution. In the Medibank matter, Deloitte reports prepared after the cyber incident were found not to be protected by legal professional privilege. The report treats the ruling as underlining the importance of how post-incident reviews are commissioned. A review commissioned the wrong way may lose the protection a company assumed it had.

What tools does the report provide?

The report is not diagnosis alone. It includes several instruments intended to help organisations assess whether their cover matches their operations:

  • A matrix of exposures
  • A broker checklist for AI risk
  • Examples of where claims uncertainty can begin
  • An 11-question insurance readiness checklist

The report also carries a client endorsement from Spriggy Co-Founder Alex Badran, who described the shift from off-the-shelf coverage to tailored solutions.

"Before working with Knightcorp, it felt like we were trying to piece together off-the-shelf solutions from different providers. Knightcorp spent the time to understand our unique needs and built insurance solutions to fit our business," Badran said.

The Forerunner Report – Technology lands at a moment when AI exclusions are entering policy wordings, regulators are demanding substantiated disclosure, and courts are testing the privilege attached to post-incident reviews. Brown's closing standard gives technology leaders the metric that matters: whether their risk and insurance arrangements still reflect the business they have become, checked before a claim forces the answer.

Source: itbrief.com.au

Share this article:

More from Rebecca Stone

Rebecca Stone

Show full bio

Correspondent covering companies, deals and strategy in artificial intelligence and machine learning at AI In Context.

622 articles

Related articles

  1. Enterprise AI Guardrails Are Shifting From Frontier Fear to Buyer Contracts
  2. OutSystems ships Agentic FNOL to automate first contact in auto claims
  3. Cowbell's New AI Liability Policy Insures Rogue Agents
  4. SAS: Corporate AI's 2027 Test Will Be Trust, Not Deployment Speed
  5. PwC survey: AI accountability scattered, 11% say it's unclear

« Previous article