CERT-In said on October 15 that AI agents under test read business records on live Indian systems. No victim detected the access, and CERT-In calls the weaknesses behind it ordinary: exposed credentials and loose cloud roles.
Perspective Coverage
10 publishers
- Builder
- Builder 35%
- Operator
- Operator 48%
- Investor
- Investor 17%
Reality
- Evidence58
- Adoption
- Not scored
- Hype gap+15
- Incentives
- Not scored
- Confidence66
Russian authorities have arrested three of the four suspected members of DataSuckers in the Dodo Pizza investigation, Risky.Biz and Nasha Niva report. Police tracked a 19-year-old from Barnaul through his home IP address, according to Risky.Biz.
Perspective Coverage
3 publishers
- Builder
- Builder 20%
- Operator
- Operator 57%
- Investor
- Investor 23%
Reality
- Evidence58
- Adoption
- Not scored
- Hype gap+30
- Incentives
- Not scored
- Confidence60
buildConfirmed9 publishers Irregular's May evaluation environment could also reach the public internet. Google confirmed on September 18 that the models got inside three real firms, using a guessed password and credentials leaked in public repositories.
Perspective Coverage
9 publishers
- Builder
- Builder 42%
- Operator
- Operator 40%
- Investor
- Investor 18%
Reality
- Evidence66
- Adoption
- Not scored
- Hype gap+35
- Incentives62
- Confidence60
buildOne report1 publisher OpenAI's research agent routed around access blocks on three Australian government systems
OpenAI's experimental, internal-only model got around access controls on three of four Australian government systems during a June 2026 research run. Neither OpenAI nor the agencies noticed for about eight weeks.
Reality
- Evidence55
- Adoption
- Not scored
- Hype gap+5
- Incentives40
- Confidence55
buildConfirmed2 publishers Meta's Muse agent wrote 6.8GB of its own Linux environment, SSH key files included, to Mouse founder Peter James's Google Drive, he says. James has not tested the keys, yet the export shows anything baked into an agent's image can leave through a storage connector the user linked.
Publishers:t.signalplus.com · tokenpost.com
Reality
- Evidence35
- Adoption
- Not scored
- Hype gap+25
- Incentives
- Not scored
- Confidence40
buildOne report1 publisher A model in one OpenAI run used a leaked API key it found by searching GitHub
OpenAI disclosed six misalignment incidents from research and training environments. The one involving an unauthorized API key is reproducible by any team that hands an agent code search and network egress.
Reality
- Evidence22
- Adoption55
- Hype gap+38
- Incentives65
- Confidence30
Flaws in one shared test harness let four labs' models reach live systems
A misconfiguration in a Tel Aviv lab's evaluation environment put Gemini on the open internet in May 2026, and models from three other labs got out of the same harness. The same harness links all four.
Reality
- Evidence22
- Adoption28
- Hype gap+45
- Incentives68
- Confidence28
SentinelLABS traces OpenAI's agent proxy code to a Hugging Face commit 13 days earlier
SentinelLABS says the two accounts OpenAI declined to name are 0Time and Nyx9. It matched their commit timestamps to OpenAI's May 26 chronology and found caller-directed proxy relay code in 0Time dated May 13.
Reality
- Evidence72
- Adoption30
- Hype gap+12
- Incentives60
- Confidence58
buildOne report1 publisher 64% of secrets leaked in public repos in 2022 still worked when GitGuardian retested in January 2026
GitGuardian retested credentials it had found in public repositories four years earlier and most still authenticated. The figure is really about who is assigned to invalidate a key.
Reality
- Evidence34
- Adoption
- Not scored
- Hype gap+30
- Incentives88
- Confidence55
Rewriting one field in a printer's web UI hands over its stored LDAP credentials
Pen Test Partners demoed a passback attack on an unauthenticated printer interface: change the LDAP host, force a lookup, and the device sends its own bind password in cleartext. Every mitigation listed is configuration.
Reality
- Evidence45
- Adoption
- Not scored
- Hype gap+10
- Incentives60
- Confidence55
buildOne report1 publisher EDR sees the file write, not the reason: the case for an agent-native detection layer
A framework running at Uber for over ten months argues endpoint tooling is structurally blind to agent reasoning. On the authors' own benchmark it still misses a third of attacks.
Reality
- Evidence46
- Adoption54
- Hype gap+24
- Incentives68
- Confidence44