AI in Cybersecurity Statistics 2026: Adoption, Costs and Threats

One in four malicious breaches were AI-enabled, a 56% increase over last year, and these breaches cost an average of $6 million, according to IBM’s 2026 Cost of a Data Breach Report. That is roughly $1 million more than the global breach average of $4.99 million. On the defense side, AI in cybersecurity statistics from the World Economic Forum (WEF) show that 77% of organizations have adopted AI for cybersecurity in its Global Cybersecurity Outlook 2026 survey.

The figures below test that adoption against depth of use, breach cost, the securing-AI market and AI-powered fraud. In IBM’s study, extensive use of AI and automation in security delivered cost savings of $1.93 million compared to organizations using none, yet one in four organizations have still not adopted these tools in their security operations.

Key Takeaways

• 77% of organizations in the WEF survey have adopted AI for cybersecurity, but organizations that used security AI and automation extensively accounted for 36% of the organizations IBM studied, up from 32% last year.
• Extensive use of AI and automation in security delivered cost savings of $1.93 million compared to organizations using none, while AI-enabled breaches cost roughly $1 million more than the global breach average of $4.99 million in IBM’s 2026 report.
• Organizations that used AI and automation extensively shortened identification and response times to 215 days, 65 days faster than those not using these technologies at all (280 days), per IBM.
• Gartner projects the market for securing AI to reach almost $4.8 billion in 2027, a 68.7% increase over 2026.
• In 2025, the FBI’s Internet Crime Complaint Center (IC3) received 22,364 complaints reporting AI-related information, with adjusted losses of $893,346,472.
• Among organizations that experienced an AI-related breach, 92% lacked proper AI access controls, according to IBM’s 2026 report.

Editor’s Choice

• AI-enabled adversary operations: up 89% year-over-year, according to CrowdStrike.
• Average eCrime breakout time: 29 minutes in 2025, with the fastest observed breakout in 27 seconds.
• Respondents assessing the security of AI tools: 64% in 2026, having nearly doubled from 37% in 2025, per the WEF.
• Security incidents involving shadow AI: 43% this year, more than doubled from 20% last year, per IBM.
• Cybersecurity professionals with AI tools integrated: 28% of respondents in ISC2’s 2025 workforce study.
• AI-linked investment fraud losses: surpassed $632 million in 2025, per the FBI’s IC3.

AI in Cybersecurity Statistics: Adoption Rates

• AI adoption inside security programs starts from a high base: 77% of organizations have adopted AI for cybersecurity, according to the WEF’s Global Cybersecurity Outlook 2026.
• Organizations that used security AI and automation extensively remained a minority but grew this year, accounting for 36% of the organizations IBM studied, up from 32% last year.
• One in four organizations have still not adopted AI and automation in their security operations, according to IBM’s 2026 breach research.
• 28% of respondents to ISC2’s 2025 Cybersecurity Workforce Study, an online survey of 16,029 cybersecurity practitioners and decision-makers, have already integrated AI tools into their operations, and in total 69% are engaged in adoption activities, whether through integration, active testing, or early evaluation.
• Adopting AI for cybersecurity and running it across the security lifecycle are different measures, and the surveys that track depth report far lower shares than the headline adoption figure.

Source: World Economic Forum Global Cybersecurity Outlook 2026, ISC2 2025 Cybersecurity Workforce Study, IBM Cost of a Data Breach Report 2026

About This Data

Compiled from 15 sources: 3 US and UK government publications and 8 company reports and newsroom posts from IBM, CrowdStrike, Google, Anthropic and Microsoft. The other 4 are analyst and industry-body publications from Gartner, ISC2 and the World Economic Forum. Publication dates run from May 2025 to October 2026. Only original research, government data and named analyst releases qualified. Figures are reviewed on a rolling basis and updated when sources publish new editions.

How Security Teams Use AI

• Organizations use AI for cybersecurity primarily to enhance phishing detection (52%) and intrusion and anomaly response (46%), in a WEF question where respondents could select up to three uses.
• AI agents already work in security operations: among organizations that have an SOC and deployed agents in it, more than 50% reported using agents for threat detection and containment, while only 18% apply agents to vulnerability management, leaving known exposures to linger even as AI shortens exploit windows, according to IBM.
• Three quarters of organizations say frontier AI threats are prompting them to rethink how agents are deployed across their security operations.
• Through analysis and automation, AI can help organizations identify vulnerabilities, detect threats, accelerate investigations, and automate security responses, according to Microsoft.

Use only matters if it shows up in breach outcomes, and IBM measures that directly.

Recent Developments

• October 13, 2026: Microsoft said that for advanced attackers, AI can even reduce attack chains that previously took days into seconds.
• October 1, 2026: Microsoft released its 2026 Digital Defense Report, finding that threat actors are incorporating AI into reconnaissance, social engineering, malware and exploit development, and post-compromise activity.
• September 8, 2026: The Google Threat Intelligence Group (GTIG) reported that in Q2 2026, GTIG observed threat actors compromise a cloud resource, then plan, build, and execute an agent-enabled mass credential harvesting campaign in under six hours.
• September 2026: Anthropic reported misuse it disrupted between December 2025 and August 2026 across seven harm areas, and identified more than 20 distinct organizations targeted in one actor’s operational planning, reconnaissance, and live operations.
• August 26, 2026: Gartner forecast the market for securing AI to reach almost $4.8 billion in 2027 and almost $7.7 billion by 2028.
• July 29, 2026: IBM’s follow-on research found 85% of organizations plan to increase security spending after becoming aware of advanced frontier AI cyber capabilities.

AI Impact on Data Breach Costs

• According to IBM’s 2026 Cost of a Data Breach Report, extensive use of AI and automation in security delivered cost savings of $1.93 million compared to organizations using none.
• Organizations that used AI and automation extensively shortened identification and response times to 215 days, 65 days faster than those not using these technologies at all (280 days).
• IBM’s global breach cost average is the baseline: AI-enabled breaches cost an average of $6 million, roughly $1 million more than the global breach average of $4.99 million.
• The $1.93 million saving from extensive security AI is almost twice the roughly $1 million premium an AI-enabled breach carries.

By the numbers: In IBM’s 2026 breach study, extensive use of AI and automation in security delivered cost savings of $1.93 million compared to organizations using none, and extensive users shortened identification and response times to 215 days against 280 days. AI-enabled breaches cost roughly $1 million more than the $4.99 million global average, so AI pushes the bill both down and up.

Gartner tracks a separate spending category built only to secure AI systems.

AI Security Market Size

• The market for securing AI is projected to reach almost $4.8 billion in 2027, a 68.7% increase over 2026, according to Gartner, per its securing AI market forecast.
• By 2028, the market is forecast to reach almost $7.7 billion.
• Gartner’s Table 1 puts total securing AI spending at $2,835 million in 2026 and $4,783 million in 2027.
• AI application security will remain the largest of the four named segments in 2027, forecast to reach almost $851 million, followed by AI usage control at $749 million.
• Other securing AI spending is forecast at $2,292 million in 2027, up from $1,368 million in 2026.
• The market for securing AI is divided into four main areas: AI application security, AI usage control, AI governance platforms, and AI gateways.
• Gartner’s figure covers tools that secure AI systems, a narrower market than all spending on AI-powered security products.

Growth rates show which segment is moving fastest.

Fastest-Growing AI Security Segments

• AI usage control will experience the largest growth of 73% in 2027, followed by AI gateway at 70.9%, according to Gartner.
• Gartner puts total securing AI spending growth at 83.0% in 2026 and 68.7% in 2027.
• AI usage control spending grows 86.6% in 2026, against 82.1% for AI application security.
• Every segment’s growth rate eases from 2026 to 2027, so the market keeps expanding at a slower pace.

On the other side of that spending, attackers already use AI across the intrusion chain.

AI-Driven Cyberattack Statistics

• AI-enabled adversaries increased operations by 89% year-over-year, weaponizing AI across reconnaissance, credential theft, and evasion, according to CrowdStrike’s 2026 Global Threat Report.
• The average eCrime breakout time fell to 29 minutes, a 65% increase in speed from 2024, with the fastest observed breakout ever occurring in just 27 seconds.
• In one intrusion, data exfiltration began within four minutes of initial access.
• Most AI-driven attacks reported in IBM’s study targeted critical infrastructure sectors (62%), with financial services and energy organizations experiencing the highest concentration.
• Deepfake impersonations drove the highest volume (45%) of AI-driven attacks in IBM’s 2026 report.
• Reported ransomware incidents rose compared to the year prior, to 39% from 34%, with attackers increasingly using AI to automate and scale, per IBM.
• Large language models (LLMs) now appear in malware: Russia-nexus FANCY BEAR deployed LLM-enabled malware (LAMEHUG) to automate reconnaissance and document collection.

Source: CrowdStrike 2026 Global Threat Report, IBM Cost of a Data Breach Report 2026

The fraud side of AI shows up in complaints filed with the FBI.

AI-Related Complaints to the FBI

• In 2025, IC3 received 22,364 complaints reporting AI-related information, with adjusted losses of $893,346,472, according to the FBI’s 2025 Internet Crime Report.
• 4,356 investment complaints and 1,764 extortion complaints referenced AI in IC3’s 2025 data.
• Phishing and spoofing complaints that referenced AI numbered 803, against 763 for harassment and stalking.
• AI technology enables the creation of convincing synthetic content, such as social media profiles and personalized conversations, often in mass quantities, the FBI says.

Complaint counts and dollar losses rank the scams differently.

AI-Linked Fraud Losses by Scam Type

• In 2025, losses in Investment complaints with a reported AI-nexus surpassed $632 million, per the FBI’s IC3.
• In business email compromise (BEC), businesses reported losses over $30 million to BEC scams involving AI in 2025.
• Employment scams involved the use of voice spoofing, or potentially voice deepfakes, during online interviews of the potential applicants.

Source: FBI Internet Crime Complaint Center 2025 Internet Crime Report

Worth noting: Overall losses to Investment scams exceeded $8 billion in 2025, while losses in Investment complaints with a reported AI-nexus surpassed $632 million, demonstrating that many victims do not realize the extent AI may be involved in scams, according to the FBI’s IC3. The AI-linked totals read best as a floor, since they count only complaints where AI was reported.

Attackers also aim at the AI systems companies now run.

Attacks on AI Systems and Shadow AI

• More than 20% of organizations reported a breach targeting AI models or applications, and the most common causes were weaknesses in surrounding systems, according to IBM.
• Among organizations that experienced an AI-related breach, the vast majority (92%) lacked proper AI access controls.
• Security incidents involving shadow AI, where workers use unapproved AI, more than doubled to 43% this year from 20% last year, and in about one in five incidents organizations reported paying a fine.
• Shadow AI incidents led to higher average breach costs this year ($5.39 million) than last year ($4.63 million).
• The global average breach cost of an AI model inversion attack was $6 million, reflecting the growing difficulty of protecting training data and sensitive information.
• Generative AI tools are now a target: adversaries are actively exploiting AI systems themselves, injecting malicious prompts into GenAI tools at more than 90 organizations and abusing AI development platforms, CrowdStrike reports.
• GTIG observed adversaries with wide-ranging motivations target proprietary AI models and source code, exfiltrate application programming interface (API) credentials, and co-opt victim cloud environments to sustain unauthorized AI workloads.
• Gartner predicts over half of successful cyberattacks on AI agents are expected to exploit access control weaknesses and prompt injections by 2029.
• The weak points sit around the models, in access controls, plug-ins and cloud settings, more than in the models themselves.

Source: IBM Cost of a Data Breach Report 2026

Readiness surveys show how many organizations check AI tools before they deploy them.

AI Security Governance and Readiness

• AI is anticipated to be the most significant driver of change in cybersecurity in the year ahead, according to 94% of survey respondents in the WEF’s Global Cybersecurity Outlook 2026.
• 87% of respondents identified AI-related vulnerabilities as the fastest-growing cyber risk over the course of 2025.
• The WEF report names insufficient knowledge or skills (54%) to deploy AI for cybersecurity as one of the main hurdles.
• Security spending plans are moving with AI: in IBM’s May 2026 follow-on research, 85% of organizations said they plan to increase security spending after becoming aware of advanced frontier AI cyber capabilities, compared to just 64% that reported in the initial research that they plan to increase security spend after experiencing a breach.
• IBM’s follow-on study was conducted in May 2026, where 456 organizations of the 602 from its breach research responded.

Key finding: The percentage of respondents assessing the security of AI tools has nearly doubled from the previous year, from 37% in 2025 to 64% in 2026, according to the World Economic Forum’s Global Cybersecurity Outlook 2026. Checking AI tools before deployment is now the majority practice, yet a large minority still skips that step.

The skills gap behind those barriers shows up in workforce data.

AI Skills in the Cybersecurity Workforce

• 73% believe AI will create more specialized cybersecurity skills, and 72% say AI will create the need for more strategic cybersecurity mindsets, in ISC2’s survey of cybersecurity professionals.
• AI remains one of the top skills needed for the second consecutive year, as 41% of respondents cited AI as a top skill needed, followed by cloud security (36%).
• Others are educating themselves on AI solutions at risk to better understand vulnerabilities and exploits (35%).
• ISC2 collected these responses in mid-2025, so they are more than a year old.

Will AI affect cybersecurity jobs?

73% of ISC2 respondents believe AI will create more specialized cybersecurity skills, and 66% say AI will require broader skillsets across the workforce. Nearly half (48%) of respondents are already working to gain more generalized AI knowledge and skills. The survey measures changing roles, not job losses; data on AI replacing jobs covers that question across occupations.

Government Guidance on AI Security

• The Cybersecurity and Infrastructure Security Agency (CISA), Australian Signals Directorate’s Australian Cyber Security Centre (ASD ACSC), and other US and international partners published Careful Adoption of Agentic Artificial Intelligence (AI) Services, a joint guide on the cybersecurity challenges and risks associated with introducing agentic AI, according to CISA.
• The guide says these systems can introduce additional cybersecurity risks, such as an expanded attack surface, privilege creep, behavioral misalignment, and obscure event records.
• Its recommendations include avoiding broad or unrestricted access, especially to sensitive data or critical systems, and beginning with agentic AI use cases that are low-risk and non-sensitive.
• The UK National Cyber Security Centre (NCSC) assessed that AI will almost certainly continue to make elements of cyber intrusion operations more effective and efficient, leading to an increase in frequency and intensity of cyber threats.
• The NCSC also judged there will almost certainly be a digital divide between systems keeping pace with AI-enabled threats and a large proportion that are more vulnerable, making cyber security at scale increasingly important to 2027 and beyond.
• Proliferation of AI-enabled cyber tools will highly likely expand access to AI-enabled intrusion capability to an expanded range of state and non-state actors, in the NCSC’s view.
• The NCSC assessment dates from May 2025, so its judgements are more than 12 months old.

Conclusion

One in four malicious breaches were AI-enabled, and these breaches cost an average of $6 million, while 77% of organizations have adopted AI for cybersecurity. The gap between those two numbers is depth: extensive use, agents on vulnerability work and access controls on AI systems all trail the headline adoption figure.

Security AI helps reduce breach cost and response time in IBM’s data, but it does not remove the risk. The AI systems companies deploy now need the same access discipline as any other critical system, and CISA’s guidance on agentic AI starts there.