Cisco

Cloud Security Engineer accounted for 9% of cybersecurity job postings across G7 economies.

Security Engineer accounted for 18% of cybersecurity job postings across G7 economies and lead other roles by share.

Cybersecurity Engineer / Analyst accounted for 16% of cybersecurity job postings across G7 economies.

Eight of 15 models have an absolute cross-regime gap greater than 15 percentage points.

Multi-turn attack success rate (ASR) ranges from 7.89% to 88.30% across the 15 closed/proprietary flagship models in the cohort.

Single-turn attack success rate (ASR) ranges from 2.19% to 64.91% across the 15 closed/proprietary flagship models in the cohort.

61% of organizations in industrial sectors are deploying AI at scale.

48% of organizations in industrial sectors identify security as their biggest networking challenge.

85% of organizations in industrial sectors expect AI to improve their cybersecurity posture.

72% of all organizations surveyed are at least moderately prepared to control and secure agentic AI systems.

84% of Pacesetters (most AI-ready group) control agent actions with guardrails and live monitoring vs 24% of all companies.

Talent gaps are most acute in the area of cybersecurity for AI, affecting 60% of all companies surveyed.

61% say secure networking is critical.

94% believe an improved network will enhance their cybersecurity posture.

One severe outage per business per year, often driven by factors including cyberattacks, adds up to $160B globally.

96% of organizations plan to upgrade their IT infrastructure.

Over 77% of organizations report that their complex security infrastructures are impeding their ability to respond swiftly and effectively to threats. These infrastructures are often dominated by the deployment of more than ten point security solutions

71% of respondents anticipate business disruptions from cyber incidents within the next 12 to 24 months.

91% of organisations trust global providers for better data protection, which is up five percentage points year-over-year.

Nearly half of respondents admit to inputting personal employee or non-public data into GenAI tools.

64% of respondents worry about inadvertently sharing sensitive information publicly or with competitors.

36% of surveyed security practitioners struggle to translate trasked data into clear business impact.

39% of security teams say critical insights are getting missed because the data lives somewhere they can’t reach.

54% of security practitioners are very confident in their ability to detect and respond to a compromised AI agent or autonomous system.

Governance, Risk, and Compliance Analyst accounted for 6% of cybersecurity job postings across G7 economies.

Cybersecurity job demand across G7 economies grew 6.8% in April–September 2025 compared to the same period a year earlier.

SOC Analyst accounted for 12% of cybersecurity job postings across G7 economies.

The share of G7 cybersecurity job postings requiring AI skills averaged 28.5% in October 2025–March 2026, up from 14.2% in October 2024–March 2025.

Cybersecurity job demand across G7 economies grew 9.5% in the period October 2025–March 2026 compared to the same six months a year earlier.

Application Security Engineer and Identity and Access Management Engineer each accounted for 7% of cybersecurity job postings across G7 economies.

Threat Intelligence Analyst accounted for 5% of cybersecurity job postings across G7 economies.

Gemini 3 Pro shifts from 18.10% single-turn ASR to 73.35% multi-turn ASR, a 4x increase.

Cross-regime deltas (multi-turn ASR minus single-turn ASR) range from −34.74 percentage points to +55.25 percentage points across the cohort.

Nova 2 Lite shows 34.05% single-turn ASR but 7.89% multi-turn ASR.

Enabling reasoning on Grok 4.1 Fast reduces multi-turn ASR from 88.30% to 43.47%.

Within each multi-turn attack strategy family, the spread between the most- and least-exposed models ranges from 79.51 to 89.25 percentage points.

Multi-turn attack success rates run 2x to 10x higher than single-turn baselines across eight open-weight LLMs in an earlier evaluation.

Imposter AI is more than 14 percentage points higher than the tenth-ranked procedure by weighted ASR.

Anthropic Claude-family single-turn ASR ranges from 2.19% to 3.64%, and multi-turn ASR ranges from 11.16% to 16.20%.

Grok 4.1 Fast in its non-reasoning configuration records a multi-turn ASR of 88.30%.

In an earlier open-weight evaluation, multi-turn ASR reached 92.78% against Mistral Large-2.

GPT-5.4 moves from 2.74% single-turn ASR to 24.68% multi-turn ASR, a 9x increase.

Imposter AI procedures produce a 37.50% weighted single-turn ASR, Soft Paraphrase produces 29.21%, and System Prompts produce 27.69%.

40% of organizations in industrial sectors cite cybersecurity concerns as a top obstacle to AI adoption.

75% of Pacesetters (the most AI-ready group) report having the capacity to fully secure agentic AI systems that make decisions independently.

55% of organizations surveyed prioritize defending models and data from tampering.

Less than a third (31%) of organizations surveyed report that they are fully equipped to control and secure agentic AI systems.

84% of Pacesetters (most AI-ready group) report full capability with end-to-end encryption for sensitive data, versus 30% of all companies.

74% of Pacesetters (most AI-ready group) report full capability with dynamic & granular access controls for AI systems and datasets, versus 27% of all companies.

Only 29% of all companies have fully integrated AI into security and identity systems, compared to 62% of Pacesetters (most AI-ready group).