If the first part of our expert predictions series showed us how fast the threat landscape is shifting, part two shows us what happens when that shift becomes structural. Across this next set of expert insights, a pattern can be seen: the attack surface is no longer something organizations “manage”, it’s something that is growing on its own.
Agentic AI is scaling identities at lightning speed, leaving governance once set by people in the dust. Meanwhile, connectors and integrations are turning into silent backdoors, Zero Trust is buckling under the weight of non-human identities, and AI-driven malefactors are operating in ways that our conventional policy models do not address.
Then there’s quantum risk (once a distant event horizon), which is now colliding with the hard reality of legacy systems that can’t simply be patched into resilience.
Another major theme is just as crucial: human judgment is becoming the last dependable control. AI has its foot on the gas on every front. Attacks are faster, automation is deeper, misinformation spreads further, and malware evolves all on its own. Yet, it’s still people who catch the things the systems miss. The little cracks, the misfires, and the quiet failures that are happening in the background.
In this next instalment, we’ll examine how these tensions meet. We live in a world where attackers move faster, defenders see more, and the line between state, criminal, and commercial actors are blurring.
It’s a reminder that 2026 won’t be defined by one breakthrough or one crisis, but by the accumulation of small, interconnected risks that paint a much larger picture.
Recognizing How Much We Rely on People
“If 2025 was the year AI flooded the enterprise, 2026 will be the year we recognize just how much we still rely on people,” says Anastasios Arampatzis, Account Manager at Bora. “As organizations continue to embrace AI and agentic systems, it becomes clearer that human oversight is not a “nice to have.” It is the critical control layer. These systems are only as safe as the judgment, domain expertise, and ethical awareness of the people guiding them. We have already witnessed AI agents “attacking” other agents, therefore, the belief that security can become fully autonomous is dangerous; every hallucination, misalignment, and unmonitored agent reminds us that human intelligence remains the ultimate failsafe.”
Arampatzis believes regulation will be the second major arena, and Europe will take center stage. “The proposed Digital Omnibus reflects a political mood shift that feels uncomfortably familiar: a pivot toward the Washington model, where innovation speed is allowed to overshadow digital rights. The growing lobbying power of Big Tech is now shaping the boundaries of cybersecurity and privacy legislation in ways that will define Europe’s digital identity for years. This is not a technical battlefield, but rather an ideological one.”
He says from a technology perspective, quantum will become the new hype cycle. “Not because quantum computing is new, but because we are finally approaching the inflection point at which “post-quantum readiness” moves from theoretical to existential. With cryptographic deadlines between 2030 and 2035, businesses cannot afford to be passive observers. Crypto-agility assessments, migration planning, and architectural redesigns must begin now (not to say yesterday).”
Meanwhile, Arampatzis says attackers will do what they always do: exploit human error, social trust, misconfigurations, and unpatched systems. “What differentiates their actions from the past is the evolution of their tooling arsenal. We must understand that it is their tools that evolve, not their playbook. In 2026, the most radical cybersecurity strategy won’t be futuristic. It will be returning to fundamentals and finally taking them seriously.”
Identity Will Become the New Ransomware
Robert Johnston, GM, Adlumin believes that identity will eclipse ransomware as the top cybersecurity battleground in 2026. “As small and mid-sized businesses continue migrating to the cloud and integrating AI applications into their workflows, many will unknowingly expose critical access points through poorly configured connectors and SaaS integrations.”
He says these identity-based attack paths, linking everything from CRM systems to cloud email, will become prime targets for threat actors looking to exploit misconfigurations and weak authentication. “What was once a simple phishing email is evolving into identity hijacking at scale, driving a new wave of breaches that begin long before encryption and ransom demands ever occur.”
Managing Non-Human Identities
Paul Davis, Field CISO at JFrog, says in 2026, Zero Trust will remain a cornerstone of security, but its implementation will become significantly more complicated adding not a replacement, but an additional burden for CISOs and security teams. “The rapid adoption of agentic AI and non-human identities is reshaping the security landscape, introducing unprecedented complexity to access management and threat detection. In fact, machine identities outnumber human identities by a factor of 45 to one on average, and in large organizations, non-human identities outnumber human users by 50 to one. What’s more, these intelligent agents often bypass traditional silos, making it increasingly difficult to enforce granular permissions and isolate access.”
He adds that as we move into the new year and beyond, developers and security leaders must contend with environments where access is not just about human credentials, but also about controlling intelligent agents whose permissions are far less transparent. “Security leaders must rethink how they verify and monitor every interaction, moving beyond legacy controls to embrace continuous authentication and real-time oversight. This includes embedding security into every phase of the software development lifecycle, leveraging continuous authentication, real-time monitoring, and automated threat detection to address risks that are no longer confined to just human users.”
Shift to Compliance as Code
As these new threats evolve, Davis says security leaders must shift to a compliance as code approach in 2026, bringing compliance standards to the business level. “This means having the tools and visibility needed to determine and demonstrate whether applications are trustworthy, confirm they meet required criteria, and being able to validate every component within their environment.”
Traditionally, Zero Trust has focused on people (and now NHIs) and IT infrastructure. Davis says we can foresee a greater focus on Attribute-Based Access Control (ABAC), or the data being leveraged in AI. “At the moment, there are few mechanisms to monitor and control data in the context of Zero Trust access within an AI service, but regulations and exposure management will drive the need for greater accountability and oversight around how data is integrated into these solutions. The organizations that will thrive are those that treat Zero Trust as a strategic enabler of innovation, not just a compliance checkbox. With this focus, security leaders can build resilient ecosystems where AI accelerates growth without compromising trust.”
The Quantum Menace
As quantum computing advances, Davis says the “Decrypt Later” dilemma will become a pressing issue for governments and enterprises alike. “While the solution may seem straightforward—simply update the software to quantum-resistant standards—the reality is far more complex. Legacy systems are often deeply intertwined with business operations, and updating them can risk data loss, incompatibility, and operational disruption. The sheer scale of technical debt across industries makes quantum resilience a complex, urgent priority that demands visionary leadership.
“Looking ahead to 2026, forward-thinking organizations must treat quantum migration as a strategic imperative, not a technical afterthought. By developing comprehensive migration plans, security leaders can help ensure that systems are resilient against both current and future attack vectors, while rigorous validation and continuous monitoring will further strengthen defenses.”
For Davis, it is those who act now who will set the standard for software security in the quantum era. “Proactive planning and decisive action will not only mitigate the risks posed by quantum-enabled threats, but also transform quantum migration into a catalyst for sustainable digital transformation and long-term resilience.”
AI’s Moment of Truth
“2026 will bring cybersecurity’s AI moment of truth,” says Michael Gray, CTO of Thrive. “The first major public breach of an AI model will expose how fragile model pipelines and data integrity really are. Attackers will learn to ransom or corrupt models used in high-stakes industries like insurance and finance, forcing companies to treat AI as critical infrastructure rather than an experimental tool.”
He says the fallout will drive investment in model governance and ‘AI firewalls’ that verify system output for accuracy and trustworthiness, which is proof that advanced defenses mean little without getting the basics right. “In the aftermath, regulators and security leaders will face growing pressure to define standards for AI integrity before public trust erodes further.”
Defensive Capabilities Will Overtake Offensive
Nicole Reineke, Senior Distinguished Product Leader, AI, says while threat actors are quickly accelerating their tactics with AI-enabled scale, defenders are poised to regain the advantage in 2026 because they can see the whole board.
“Unlike attackers, who often operate alone, with limited creativity, security vendors can aggregate patterns across thousands of attempted intrusions to better understand popular tactics and strategies. This cross-actor visibility allows defenders to proactively identify emerging techniques long before individual organizations are targeted. In 2026, this network-level intelligence will become one of the most powerful differentiators in cyber resilience, enabling defenders to predict and neutralize attacks before they begin.”
The Next Great Security Gap
The biggest data breaches of 2026 will come from the tools companies willingly connect to their own systems, comments Robert Johnston, GM of Adlumin. “As AI apps proliferate across cloud ecosystems, businesses will rush to integrate them without understanding the permissions and data flows involved. Each connector becomes a potential backdoor into business-critical systems like CRMs, SaaS platforms, and email environments. Attackers will exploit these weak links faster than defenders can catalogue them, forcing the industry to evolve more quickly toward disciplines like AI Posture Management.”
Eroding Critical Thinking Skills
Widespread use of AI-driven information tools will make it easier than ever to find answers – but at the cost of eroding critical thinking skills, says David Higgins, Senior Director – Field Technology Office EMEA, at CyberArk. “As people increasingly rely on AI to provide instant responses, the imperative to evaluate sources and exercise judgment will diminish. This creates fertile ground for social engineering and misinformation campaigns, as malicious actors can flood the internet with false narratives that AI systems may inadvertently amplify.”
Higgins says the risk is not just technical, but societal. “As users become less accustomed to questioning the validity of information, organizations will face new challenges in defending against manipulation and accidental breaches caused by misplaced trust in AI-generated content.”
The Vulnerabilities of Interconnected Systems
Cyberark’s Director, EMEA, financial services and insurance, Andy Parsons, adds that as leading banks strive to lower their cost of capital, they increasingly rely on fintech partnerships and digital payment innovations to expand their reach and attract more deposits.
“This interconnected ecosystem, built for efficiency and growth, unintentionally creates new security vulnerabilities. Each additional fintech or digital payment provider becomes a potential entry point for attackers, expanding the attack surface and introducing risks such as identity theft, synthetic fraud, and disruption of payment systems. The drive for lower costs and greater reach, while commercially compelling, can outpace the ability of banks and their partners to secure every link in the chain, making the entire financial system more susceptible to sophisticated cyber threats.”
The Year of Salt
China’s cyber operations have reached full-spectrum maturity. Volt Typhoon revealed long-term “living off the land” access to U.S. critical infrastructure, adds Steve Stone, SVP, Threat Discovery & Response at SentinelOne. “At the same time, the newly exposed Salt Typhoon campaign (spanning over 80 global telecoms) demonstrated the scale and precision of state-backed espionage. Together, they form bookends for China’s modern cyber doctrine: persistent access now, disruption later. If Beijing moves on Taiwan, those footholds could instantly pivot into multi-domain attacks, meaning telecom blackouts, supply-chain paralysis, maritime disruption, and destructive wiper operations designed to slow military response and destabilise civilian life.”
Stone says by 2026, the world will see the consequences of a decade of pre-positioning: a cyber battlefield already built inside global infrastructure. “Communications outages, semiconductor shocks, and AI-driven disinformation will define the first phase of any conflict. For governments and enterprises alike, resilience must be built before the storm, not after it starts.”
State-Backed Cybercrime Will Look Like a Day Job
He adds that the line between criminal enterprise and state agenda is dissolving. “SentinelLabs research into North Korea’s IT worker network revealed hundreds of front companies, many operating out of China, and over 1,000 job applications from fake DPRK-linked personas attempting to infiltrate even major cybersecurity firms. These operations show how Pyongyang’s cyber workforce now blends sanctioned espionage with commercial freelancing, using legitimate hiring pipelines to fund the regime’s illicit programmes.”
Stone believes that soon we’ll see this model become the playbook for state-sponsored revenue generation: cyber operators posing as freelancers, consultants, or contractors across global tech ecosystems to quietly fund military and intelligence operations. “The convergence of cybercrime and statecraft means the next “insider threat” may not be an employee gone rogue, but a foreign government’s operative disguised as your next remote hire.”
One of the Most Profitable Attack Surfaces
According to Stone, the group known as Scattered Spider, part of the broader collective The Com, has redefined what modern cybercrime looks like. “Built around deception rather than exploits, the collective’s members use social engineering, vishing, and help-desk impersonation to compromise major organizations across telecom, retail, and now aviation and insurance.
“Their attacks have caused hundreds of millions in damages, leveraging stolen credentials, remote tools like AnyDesk and ScreenConnect, and alliances with ransomware groups such as ALPHV/BlackCat. Despite arrests, their decentralised, franchise-like structure keeps operations alive and adaptable. Over the next few years, The Com and its offshoots will evolve into AI-enhanced social engineering networks, blending voice cloning, deepfake video, and LLM tooling to automate pretext creation and expand their reach. Expect a new wave of “human-first” cyberattacks meaning scalable, personalised, and nearly indistinguishable from legitimate communication, making the employee, not the endpoint, the real zero-day.”
Information Security Buzz News Editor
Kirsten Doyle has been in the technology journalism and editing space for nearly 24 years, during which time she has developed a great love for all aspects of technology, as well as words themselves. Her experience spans B2B tech, with a lot of focus on cybersecurity, cloud, enterprise, digital transformation, and data centre. Her specialties are in news, thought leadership, features, white papers, and PR writing, and she is an experienced editor for both print and online publications.
The opinions expressed in this post belong to the individual contributors and do not necessarily reflect the views of Information Security Buzz.