Amazon SES has added a new field to its engagement data that helps senders identify non-human interactions (NHI), such as bot activity, and separate them from genuine recipient opens and clicks.
AWS the change was announced on 7 August 2026. Open and Click event notifications can now carry an isBotEvent field, with a value of either Likely or Unlikely.
Designed to help SES users distinguish human engagement from activity generated by automated systems. That includes email security tools, link scanners and other machine processes that open messages and follow links before, or instead of, the recipient.
The non-human problem is one the security industry has long labelled non-human identities and sometimes in the email world you will hear non human interactions. Security systems fetch tracking pixels and follow links as part of routine scanning. Each fetch registers as an open. Each followed link registers as a click. No human is involved. If NHI is new territory for you, or you want a refresher on where these phantom opens and clicks come from, our video on non-human interactions and mitigating bot click activity presented by Jakub Olexa of Omnivery covers the sources in detail, from Apple MPP and security scanning to campaign archives and commerce bots, along with practical mitigation steps.
AWS is careful with its language here, and senders should be too. The classification is a likelihood signal, not a definitive determination. The developer guide documentation for event data published to Amazon SNS and Amazon Data Firehose defines isBotEvent as the likelihood that an event was generated by an automated system rather than a human.
There are no new event types. SES continues to publish standard Open and Click events. The isBotEvent field simply rides along in the event data.
For customers already publishing Open or Click events through an SES configuration set event destination, the new field arrives automatically. No changes are needed. The feature is available in all AWS Regions where Amazon SES operates.
What difference does it make?
NHI has become one of the defining measurement problems in email. Automated opens and clicks inflate engagement metrics, corrupt A/B tests and can trigger automation workflows built on the assumption that a click means a person. Senders who treat every open or click as human intent are working from bad data.
The new SES field gives senders another data point for filtering likely automated interactions out of their event processing. It brings SES into line with a growing number of platforms that surface bot signals to their customers, though approaches and accuracy vary widely across the industry.
One caveat worth noting. AWS has not detailed its detection methodology, and no accuracy rate for the classification was published with the announcement. Senders should treat the field as a useful input, not a source of truth.
