October 2026 edition · · 7 min read

86% of Firms Hit by Hiring Fraud, 67.6M Records Breached: October's Twin Crisis

Oracle Health's 20-million-patient exposure, a record 1,400-plus-CVE Patch Tuesday, Checkr's finding that 86% of organisations met hiring fraud this year, and 38% of managers suspecting a fraudster is already on the payroll. October makes the case that cybersecurity and workforce integrity are one risk conversation. From iChek (background verification) and iSec (information security).

86%

of organisations experienced hiring fraud in the past year (Checkr Hiring Hoax 2026, 3,000 US managers)

42%

of identified candidate-fraud cases escaped pre-hire detection (HYPR 2026)

22%

of recruiters in Pin's own 2026 survey confirmed a deepfake or identity swap in a video interview

1,400+

CVEs in Microsoft's October 2026 Patch Tuesday, a record release (trackers count 1,411 to 1,415)

20M

patient records reported exposed in the Oracle Health (Cerner) breach (Texas Attorney General filing)

67.6M

confirmed records breached in October 2026, a floor counted across 17 of the 49 incidents GRC Solutions tracked

Background check: hiring risk

Global

AI fraud

Checkr 2026 (3,000 US managers): 40% interviewed a fake-identity candidate, up from 31% last year

86% of organisations experienced at least one type of hiring fraud in the past year. 40% interviewed a candidate later found to be using a fake identity. 38% suspect a fraudulent employee is on their payroll right now. 28% admitted advancing a candidate they suspected of dishonesty because the role needed filling urgently.

Source: Checkr, Hiring Hoax 2026, online survey of 3,000 US managers, 11 to 16 Sep 2026. Self-reported, US only.

Deepfake detection

Pin 2026: 22% of surveyed recruiters confirmed a deepfake in a video interview

In a survey of recruiters using Pin's platform, 22.4% said they had confirmed a deepfake or identity swap in a video interview, and 16.3% more suspected one but could not prove it. It is a vendor survey of its own users, not a measure of all hiring.

Source: Pin, 2026 deepfake candidate report. Self-reported survey of Pin's own users.

Post-hire risk

HYPR 2026: 98% of HR leaders have faced candidate fraud, and 42% of cases slip past pre-hire checks

HYPR's State of HR Identity Fraud Detection report finds near-universal exposure: 98% of HR leaders have experienced candidate fraud firsthand, and in 42% of cases the fraud escaped pre-hire detection. Once it escaped, detection was slow: fewer than 3% were flagged the same day, 45% took four to six days, and one in five went undetected for up to three weeks.

Source: HYPR, State of HR Identity Fraud Detection, reported by Biometric Update, Sep 2026.

India

Sector alert

Healthcare and pharma: fake-degree cases keep surfacing in the highest-stakes sector

Indian police have recently busted rackets issuing fake pharmacy (B.Pharma) and medical (MBBS) credentials. Where a false qualification is a potential patient-safety incident and not only a contract issue, hospitals and pharma employers should verify degrees and registrations directly with the issuing university or council, including for lateral hires.

Source: news reports in Medical Dialogues. Individual cases; no trend rate is claimed.

iChek insight: the window after the hire is your biggest risk

When a fraudulent candidate gets past pre-hire screening, the damage happens before anyone notices. In HYPR's data, 45% of such cases took four to six days to detect and one in five took up to three weeks. In that window a person can copy data, set up persistence, or cause damage that takes months to repair. Continuous post-hire monitoring (periodic re-verification, UAN dual-employment audits, behavioural risk flags and social-media risk mapping) closes the gap that a one-time check leaves.

Information security: threat watch

Global

Major breach

Oracle Health (Cerner): about 20 million patient records reported exposed

A Texas Attorney General filing reported that roughly 20 million patient records were affected in the Oracle Health (Cerner) data breach. How the attackers got in has not been confirmed in the sources we could verify, so treat any root-cause claim with caution. Organisations on Oracle Health or legacy health-record platforms should review credential hygiene and legacy-system access.

Source: Texas Attorney General filing, as reported by eSecurity Planet and SecurityWeek, Oct 2026.

AI supply chain

Malicious LiteLLM 1.107.0 published to PyPI and official Docker images, carrying a credential stealer

LiteLLM said the malicious release was published through three engineers' GitHub accounts, apparently using stolen GitHub CLI OAuth tokens. The package carried a credential stealer and a backdoor, and the PyPI package and Docker images were pulled. LiteLLM had not established how the tokens were obtained. It is at least the second LiteLLM compromise of 2026, and it shows that AI toolchains need the same supply-chain rigour as production dependencies.

Source: The Hacker News, Oct 2026.

Critical CVE

PHP core remote-code-execution flaws (CVE-2026-109618 and CVE-2026-109619) emergency-patched

The PHP project shipped emergency patches for two critical remote-code-execution flaws, reportedly after details leaked ahead of scheduled disclosure. Check the PHP version on every application you run, apply the patched release, and use WAF rules as a stop-gap.

Source: BleepingComputer; Health-ISAC bulletin, 21 Oct 2026.

Record patch day

Microsoft's October Patch Tuesday: a record release of about 1,400 CVEs, with three zero-days

Trackers count 1,411 to 1,415 CVEs and three zero-days, one of them a Windows Shell flaw reported as exploited before the patch. Volume like this makes manual patch prioritisation unworkable at enterprise scale, so risk-based patch management is now a baseline.

Source: Microsoft Security Update Guide, 13 Oct 2026. Trackers differ slightly on the count: BleepingComputer 1,411, Daily Security Review 1,413, Security Affairs 1,415.

India

India alert

SBI confirms 3,897 customer records in a roughly 375,000-record dark-web dump

In an October 21 stock-exchange filing, SBI said 3,897 of its customer records appear in a dark-web dataset of roughly 375,000 rows posted by a threat actor. The source of the data is not yet known, and SBI said account balances and funds are safe. The remaining records have not been tied to SBI. The case is a reminder to check what customer data sits with partners and aggregators.

Source: DT Next and Breached.Company, Oct 2026. Origin unconfirmed.

Certificate crisis

eMudhra intermediate CA certificates revoked globally after weak private keys were found

The private keys of eight eMudhra intermediate CAs (six for TLS, two for code signing) were recoverable from their public certificates because of weak randomness in key generation. Google, Mozilla and Cloudflare blocked them. eMudhra said none of the eight had issued end-entity certificates. Organisations should confirm their trust stores have the updates.

Source: CSO Online, Oct 2026.

CERT-In

CERT-In guidance sets a 12-hour patch target for exploited, internet-facing flaws

CERT-In's guidance, published in May 2026, recommends patching known-exploited vulnerabilities on internet-facing systems within 12 hours where feasible, other critical exposed flaws within a day, and high-severity ones within five days, citing AI-assisted exploitation. Separately, the existing requirement to report cyber incidents to CERT-In within 6 hours still applies.

Source: The Hacker News, 26 May 2026. Binding force of the 12-hour target is unclear.

iSec advisory: non-human identities are your invisible attack surface

The ConnectWise Cloud Backup compromise (a stolen application credential used to reach at least 162 Microsoft 365 tenants, per Huntress) and the LiteLLM attack (stolen developer OAuth tokens) both turned on credentials held by software and not by people. Service accounts, application tokens and API keys are often over-permissioned, unrotated and outside MFA and identity governance. Rotate and scope every token and service account, and treat the secrets manager as part of your attack surface.

The compliance clock

DPDP Act

DPDP consent managers go live in November 2026; the main duties follow on 13 May 2027

The consent-manager provisions of the DPDP Rules take effect in November 2026, one year after the Rules were notified. Consent managers are an option for candidates and not a requirement for employers, but they will change how consent is given and withdrawn, so background-verification programmes should be ready to record that consent was collected digitally, covers only the checks actually run, and can be audited. Notice, security and breach-reporting duties for data fiduciaries begin on 13 May 2027. Manual or email-only consent trails will be hard to defend by then.

Source: DPDP Rules 2025 commencement schedule; see Ampcus Cyber and KS&DK on the consent-manager framework.

What's new from iChek and iSec

iSec · now available

VulnScanAI

Nmap, OWASP ZAP and leading VAPT tools in one platform. Every finding is re-analysed by AI, PII is masked automatically, and agents cover Windows, Linux and macOS. DPDP, RBI, SEBI, CERT-In and ISO 27001 reporting is built in. A free version is available.

Try VulnScanAI free →
iChek · post-hire

Continuous monitoring after the offer

Periodic re-verification, UAN dual-employment audits, behavioural risk flags and social-media risk mapping, to cover the gap that one-time pre-hire checks leave.

Employee screening →
iSec · consulting

VAPT, CERT-In and DPDP readiness

Web, mobile and network VAPT, ISO 27001 consulting, 6-hour CERT-In reporting readiness, Zero Trust architecture and incident response. iSec's VAPT and Zero Trust work now includes non-human identity assessment.

About iSec →

Sources and notes

Every figure carries the source, date or period it refers to. Published 7 November 2026. Survey figures are self-reported and describe the respondents, not all hiring.

Hiring: Checkr, Hiring Hoax 2026 (3,000 US managers, 11 to 16 Sep 2026); HYPR, State of HR Identity Fraud Detection (Sep 2026); Pin, 2026 deepfake candidate report.

Security: Texas Attorney General filing on Oracle Health, via eSecurity Planet and SecurityWeek (Oct 2026); Microsoft Security Update Guide, 13 Oct 2026; GRC Solutions monthly breach count (Oct 2026); BleepingComputer, The Hacker News, CSO Online, Help Net Security (Oct 2026); SBI filing of 21 Oct 2026; CERT-In guidance (May 2026).

Compliance: DPDP Rules 2025 commencement schedule; Ampcus Cyber and KS&DK commentary on the consent-manager framework.

Get the next issue

One short, sourced brief a month on hiring fraud, background verification and information security in India.

Email us to subscribe

Also: RSS feed · LinkedIn

← September 2026 All issues

Verify the person before the access

Background checks, identity and address APIs, and field verification, with prices published.