The cyber situation in 2026
The 2026 threat environment is shaped by three forces — more capable cybercriminals, new external shocks, and governance that hasn't kept pace. Ransomware and supply-chain attacks are escalating, while AI and geopolitical tension act as accelerants. AI is becoming a broad-spectrum threat multiplier. Since ChatGPT's launch in November 2022, malicious phishing messages have surged 4,151 percent. Voice-phishing attacks leveraging AI voice cloning rose 442 percent in 2024. Deepfake incidents multiplied roughly 1,500 percent since 2023.
Meanwhile, geopolitical actors, China, Russia, North Korea, and Iran (the CRINK alignment documented by NATO) have incorporated AI capabilities into state- sponsored campaigns.
Boardrooms have taken notice. Gartner surveys indicate 85 percent of CEOs identify cybersecurity as a critical business risk, while 90 percent of non-executive directors lack confidence in measuring cyber value. The regulatory environment is also lacking. While progress has been made with DORA and the CRA, NIS2 has been transposed by only fourteen of twenty-seven EU member states as of February 2026.22 The EU Artificial Intelligence Act requires transparency in automated decision systems, but its transposition into national laws may still remain complex.
Why video communications are uniquely exposed
Videoconferencing is considered “high priority” because it combines (a high-trust channel (invitations/participants) that does not yet incorporate the years of security culture of emails; (ii) a sense of urgency (an imminent meeting) that can further erode that security culture; (iii) high-impact native capabilities (screen sharing, files, links, integrations) that then provides rich targets; and (iv) a structural dependence on cloud identity and on multi- tenant configurations that are sometimes permissive by default, especially with external users.
These threats can be best evaluated along the Confidential / Integrity / Availability dimensions of cybersecurity, along with privacy.
Confidentiality risk is unique: video conferening is the only platform where leadership voluntarily discloses strategic intentions, competitive assessments, and financial projections in real time. Unlike asynchronous email or static repositories, it captures the unguarded moment—the concession not yet formalized, the acquisition target not yet announced.
Integrity is equally threatened: deepfake synthetic media now enables adversaries to fabricate executive identities during live calls, as the Arup, Ferrari, and WPP incidents proved. When a CFO cannot trust that the CEO on screen is genuine, the modality designed to accelerate decisions becomes an instrument of fraud.
Availability compounds the stakes: during crises—natural disasters, cyberattack responses, geopolitical emergencies—video collaboration is irreplaceable. A platform compromise at that moment turns the system meant to manage the emergency into a casualty of it.
Privacy exposure is severe. HIPAA in the US mandates encryption for protected health information; GDPR in EU imposes heightened safeguards for special-category data—health records, legal proceedings, financial situations—processed during video communications. NIS2, DORA, and the CRA extend obligations directly to communication infrastructure, with personal liability for executives under NIS2 and institutional fines under GDPR.
What is needed for this new age
Goldstein sets out the criteria a platform must now meet — software-based and rapidly deployable, integrated with zero-trust, connection-aware, running on-premise anti-deepfake detection, with highest-standard post-quantum-ready encryption, white-label deployment, RBAC/ABAC, genuine interoperability, and sovereign deployment backed by independent certification and a proven track record.
"Without connection-aware security posture assessment, even the most robust encryption becomes operationally meaningless."
