When a customer claims your product, model, or implementation caused them financial harm, Tech E&O pays the defense and the damages.
Technology Errors & Omissions — also called tech professional liability — covers financial loss a customer suffers because your technology or services failed to do what you said they would. A missed SLA, a bug that corrupted records, an integration that silently dropped transactions, a model that produced an output someone relied on. General Liability will not respond to any of that, because there is no bodily injury and no physical property damage.
It is the policy enterprise procurement is actually asking about when an MSA says 'professional liability' or 'errors and omissions.' Limits of $1M to $5M are the most common contractual asks, and the requirement usually appears alongside cyber, general liability, workers' compensation, and umbrella in the same insurance exhibit.
For AI companies, this is the coverage that has changed the most. Some carriers now attach broad exclusions for claims arising from artificial intelligence; others offer affirmative AI coverage. Two policies with the same limit and premium can behave in completely opposite ways when a model output is at the center of the claim. Reading that language is the job.
Buy it when
Grants vary by carrier and form. These are the components we look for when we place it.
Defense and damages when your product or professional services fail to perform as represented and a customer suffers financial loss.
Many forms extend to contractual liability arising from failure to meet stated performance obligations — a critical extension for SLA-driven businesses.
Copyright, trade dress, and often trademark claims arising from your product or content. Patent is typically excluded.
Where a software failure leads to physical harm — important for devtools embedded in operational, industrial, healthcare, or automotive contexts.
Claims arising from work performed on your behalf by contractors and outsourced developers.
Some forms fund the expense of fixing the failure to mitigate a larger loss. This is a differentiator worth asking for by name.
Where available, explicit coverage for claims arising from AI and machine learning outputs rather than silence or an exclusion.
We put this in front of you at binding — the only moment you can still do something about it.
Composite scenarios drawn from how these losses typically develop. Illustrative, not case files.
A release changes rounding logic in a billing platform. The customer under-collects for six weeks and cannot claw it back from consumers. They demand $1.9M. Tech E&O defends and negotiates the resolution.
A forecasting product materially misstates demand after a training data pipeline fails silently. The customer over-purchases inventory and sues for the write-down. Whether this claim is paid or denied comes down to whether the policy has an AI exclusion.
A 31-hour outage breaches an uptime commitment. The customer terminates, demands a refund of prepaid fees, and claims consequential damages from their own downtime. The professional liability form responds where the cyber trigger does not.
Ranges we commonly see for technology companies. Your contracts and exposure decide the answer — this is where the conversation starts, not where it ends.
| Stage | Typical | |
|---|---|---|
| Pre-seed / Seed | $1M | Combined Tech E&O + Cyber. Sufficient for most early SMB and mid-market contracts. |
| Series A | $2M – $3M | Enterprise MSAs commonly specify $2M. Confirm the contract counts a shared E&O/cyber limit as compliant. |
| Series B | $3M – $5M | Contractual liability wording and IP infringement scope become the real negotiation. |
| Series C+ | $5M – $20M+ | Structured towers, per-customer requirements, and named-customer endorsements. |
What we need to quote
Most of this is collected once in the Atlas submission and reused across the markets we approach.
Start a submissionNo. General Liability covers bodily injury and physical property damage. A software defect that costs a customer money is neither. Nearly every claim a software company faces falls outside GL entirely.
Almost never. Patent is excluded on standard forms. Copyright and trade dress are usually included. If patent exposure is material, that is a separate specialty market and a separate conversation.
That is a normal ask and usually solvable — either with a primary plus excess structure, or by negotiating the exhibit down to what is proportionate to the contract value. We do that negotiation with procurement regularly.
Materially. Carriers are split between silence, exclusion, and affirmative grant. We read the AI language on the quotes we present and tell you plainly which of the three you're being offered.
Compare cyber insurance and technology E&O, see where claims overlap, and learn what to check in a combined policy for a software business.
Read the guide: Cyber Insurance vs. Tech E&O: What Is the Difference?Build a SaaS insurance checklist around customer contracts, data, employees and funding. Know what to prepare before requesting quotes.
Read the guide: SaaS Insurance Checklist for Founders and Finance TeamsReview an MSA insurance exhibit: coverage types, limits, additional insured wording, certificates, endorsements and timing before you sign.
Read the guide: Enterprise Contract Insurance Requirements: A Tech Company ChecklistTell us about the business once. A broker reviews it the same business day and comes back with a plan and a timeline.
