buildConfirmed3 publishers Zenity Labs says one prompt to a public Bedrock AgentCore agent returned credentials for a role scoped to every agent in the account and region. AWS had removed the wider default permissions by September 29, 2026, but roles a team copied or widened sit in its own account.
Perspective Coverage
4 publishers
- Builder
- Builder 41%
- Operator
- Operator 41%
- Investor
- Investor 18%
Reality
- Evidence55
- Adoption
- Not scored
- Hype gap+30
- Incentives60
- Confidence58
buildOne report1 publisher ShipStation reports its Deckhand agent resolved 97.5% of help desk channel queries
ShipStation Global says its Slack IT agent, Deckhand, resolved 97.5% of incoming channel queries and drew 96% positive feedback. Both figures are self-reported from one help desk channel, so they transfer only where answers already exist in writing.
Reality
- Evidence35
- Adoption25
- Hype gap+30
- Incentives75
- Confidence70
Zenity Labs shows one prompt could pull other agents' code out of Amazon Bedrock AgentCore
Zenity Labs researchers found that a single prompt could steal temporary credentials from Amazon Bedrock AgentCore through its IMDSv1 metadata service. SC World, citing The Register, says an excess-permissions problem outlasted AWS's metadata fix.
Reality
- Evidence45
- Adoption
- Not scored
- Hype gap−5
- Incentives
- Not scored
- Confidence50
SailPoint announced conditional just-in-time access, agent discovery and an agent kill switch at its Navigate conference in Austin. Futurum's analyst calls the bet well positioned but flags runtime authorization and agent coverage as open.
Perspective Coverage
3 publishers
- Builder
- Builder 25%
- Operator
- Operator 53%
- Investor
- Investor 22%
Reality
- Evidence50
- Adoption
- Not scored
- Hype gap+20
- Incentives55
- Confidence62
AWS bundles NVIDIA's robotics tools into an open-source path from simulator to factory floor
AWS released an open-source Physical AI Toolchain that joins four AWS services and four NVIDIA tools into one simulation-to-robot pipeline. It handles the plumbing between stages, though Interesting Engineering notes it cannot guarantee a simulation-trained model works on a real factory floor.
Reality
- Evidence40
- Adoption
- Not scored
- Hype gap+15
- Incentives60
- Confidence45
RBA submissions find chargeback liability unclear when AI agents exceed their authority
Merchants, issuers and payment firms among 75 RBA submitters said chargeback liability is unclear when an AI agent exceeds its authority. Amazon's AgentCore Payments already lets agents pay in stablecoins, so for now merchants and networks are left to price the risk of a disputed order.
Reality
- Evidence50
- Adoption20
- Hype gap+25
- Incentives70
- Confidence45
AWS pushes four agent-focused launches in 19 days, from a reworked runtime to an open-source sandbox
Amazon Web Services shipped four AI-agent infrastructure launches in 19 days, among them a managed agent service built with OpenAI. Most are previews or free code, so what customers will pay for the agent layer is still unknown.
Reality
- Evidence40
- Adoption
- Not scored
- Hype gap+15
- Incentives
- Not scored
- Confidence45
buildConfirmed2 publishers Zenity Labs found that one prompt to a single public Amazon Bedrock AgentCore agent could take over every agent in the same AWS account and region. Zenity says the partial fix from AWS still leaves teams to cut each agent's role down to minimal access by hand.
Reality
- Evidence50
- Adoption
- Not scored
- Hype gap+15
- Incentives65
- Confidence55
buildOne report1 publisher AWS AgentCore Runtime Instances swap per-call cold starts for one always-on GPU host
AWS's AgentCore Runtime Instances put three agents and 18 GB of models on one 24 GB A10G host with a shared, persistent disk. The design cuts cold starts and S3 handoffs on a single host; teams still budget GPU memory themselves and pay for the hours the host sits idle.
Reality
- Evidence45
- Adoption
- Not scored
- Hype gap+5
- Incentives
- Not scored
- Confidence40
buildOne report1 publisher AWS adds seven objectives to the AI Practitioner exam guide five weeks after version 1.0
AWS published version 1.1 of the AIF-C01 AI Practitioner exam guide on April 30, five weeks after 1.0, with seven new objectives including token pricing. A dev.to review finds older courses still cover most of the exam but are thin on agents, token cost and grounding.
Reality
- Evidence55
- Adoption
- Not scored
- Hype gap+5
- Incentives60
- Confidence55
buildOne report1 publisher AWS's AgentCore SDK needed two releases to close one injection bug in install_packages()
AWS's Bedrock AgentCore Python SDK shipped fixes in v1.6.1 and v1.18.1 for one argument-injection flaw in install_packages(). Teams that let agents or users name packages for Code Interpreter sandboxes should check those names before the SDK sees them.
Reality
- Evidence55
- Adoption
- Not scored
- Hype gap+10
- Incentives
- Not scored
- Confidence50
buildConfirmed2 publishers AWS's CloudWatch Omni, generally available since September 23, lets Okta and Entra ID users investigate incidents without AWS console access. CloudWatch dashboard sharing has let outsiders view prebuilt graphs since 2020, so what Omni adds is the investigation itself, one of the reasons teams paid for third-party platforms.
Reality
- Evidence50
- Adoption
- Not scored
- Hype gap+20
- Incentives60
- Confidence55
buildConfirmed2 publishers AWS published a Bedrock AgentCore sample for agents that wake on S3 and scheduled events, capping each agent turn at Lambda's 15-minute timeout. Reviewers can take hours, so the sample stops the agent at every human checkpoint and restarts it from saved state.
Reality
- Evidence60
- Adoption
- Not scored
- Hype gap+20
- Incentives70
- Confidence65
buildOne report1 publisher Cohesity says rolling back an AI agent leaves its sent emails and outside changes in place
Cohesity's Agent Resilience rolls Bedrock AI agents back to immutable recovery points, yet its product leads say emails already sent stay sent. Outside systems still need their own clean-up and tests, and the company calls its recovery-time edge intended, not measured.
Reality
- Evidence35
- Adoption
- Not scored
- Hype gap0
- Incentives70
- Confidence40
JFrog has documented an on-behalf-of exchange so agent calls reach Artifactory as the signed-in user. The price is the Gateway's cached tool search, which per-user discovery gives up.
Reality
- Evidence55
- Adoption
- Not scored
- Hype gap+15
- Incentives70
- Confidence60
buildOne report1 publisher AWS puts the agent runtime behind an n8n node, and leaves the permissions with you
An open-source AgentCore node moves memory, sandboxed execution and tool access into a managed harness. The install is the easy part; the access decisions do not move.
Reality
- Evidence45
- Adoption
- Not scored
- Hype gap+10
- Incentives70
- Confidence50
buildOne report1 publisher AWS wraps five quality layers around the LLM assistant used by over 4,000 of its leaders
AWS says NarrateAI, its data assistant for over 4,000 executive leaders, reaches about 99 percent numerical accuracy through five layered techniques. All five sit outside the model, so copying the design means extra AWS accounts and parallel evaluators on every paragraph.
Reality
- Evidence35
- Adoption30
- Hype gap+20
- Incentives75
- Confidence40
buildOne report1 publisher HEMA wired its existing service catalog into Kiro and Claude as MCP tools
The AWS post co-written with HEMA says the structured half of its internal knowledge was already in good shape, and that the procedural half had little written documentation to fall back on.
Reality
- Evidence38
- Adoption42
- Hype gap+22
- Incentives78
- Confidence58
buildOne report1 publisher Strands Evals scores skill selection and invocation as pass/fail, rates instruction following on a five-level scale
AWS has added three skill-focused evaluators to Strands Evals and Bedrock AgentCore Evaluations. Two of them call a model once per invoked skill, and the third is a deterministic check that exists only in Strands.
Reality
- Evidence52
- Adoption10
- Hype gap+15
- Incentives80
- Confidence55
buildOne report1 publisher Trane put a tool gateway between its HVAC agent and telemetry from millions of assets
AWS says Trane's engineers cut a 20-minute dashboard workflow to a 20-second question in three to four weeks. The multiple rests on timings Trane ran with its own technicians, and the durable part is where the tool calls execute.
Reality
- Evidence34
- Adoption26
- Hype gap+32
- Incentives82
- Confidence56