What cybersecurity experts predicted for 2025 - and what actually happened
By the end of December, inboxes (or at least mine) are flooded with cybersecurity professionals hawking their predictions for the year ahead. But before we rush into 2026, how accurate were last year’s expert forecasts?
Image by yuda chen | Shutterstock
By the end of December, inboxes (or at least mine) are flooded with cybersecurity professionals hawking their predictions for the year ahead. But before we rush into 2026, how accurate were last year’s expert forecasts?
From ransomware and SaaS risk to AI-driven attacks, this cybersecurity journalist went way back in her archives to compare "hype vs reality" in 2025, examining which expert forecasts got it right and which predictions, if any, never fully materialized.
Whether cautious, bold, or speculative, I received just over 90 distinct predictions from 36 experts across the cybersecurity sector, including vendors, consultants, researchers, think tanks, and the like.
(Now, keep in mind, this is just a snapshot from my own inbox, and not a full representation of the entire security industry. But still revealing nonetheless.)
More from Cybernews
Sainsbury’s installs more facial rec tech in stores: Should shoppers be worried?
"Phantom squatting” uses AI hallucinated domains for cyber attacks
What they’re doing to gamers is a robbery in broad daylight
Irish parliament expands Microsoft use, despite the EU’s efforts to ditch it
Can government AI actually scrub UAP footage from the internet?
The EU and Apple CEO Tim Cook held "constructive" talks after their Siri AI dispute in Europe
Based on the analysis (thanks to a bit of help from AI), several themes rose to the top – not just in frequency, but in how closely they mirrored reality. The most common predictions, it seems, weren’t about novel threats, but about existing attacks becoming easier, faster, and more scalable, largely due to AI.
The consensus further appeared to be that cybersecurity in 2025 would be defined by data, automation, and attacker efficiency, rather than by entirely new threat categories.
The patterns emerge
Initially, I grouped the predictions into three distinct categories: Cybersecurity, AI & Machine Learning, and Regulation, Compliance & National Security.
However, in the end, there was such a crossover and emphasis on the overall threat landscape that I decided to create a top 10 list, ranked by how often each prediction appeared across expert submissions and the patterns that emerged.
Here are the top 10 most common cybersecurity predictions for 2025, along with the percentages of experts who submitted them.
- AI enabling more sophisticated cyberattacks (58%)
- AI shifting from experimentation to real-world use (35%)
- Ransomware landscape fragmenting (29%)
- Supply chain attacks across SaaS, cloud, OSS (29%)
- SaaS security gaps (26%)
- Zero Trust becoming mandatory (23%)
- Data protection & governance as core priorities (23%)
- Automation needed to keep pace with attackers (23%)
- Regulation increasing complexity, not stopping attacks (19%)
- Deepfakes & synthetic identity fraud (16%)
AI-enabled cyberattacks for the win
The single most dominant theme in 2025 expert forecasts was AI’s role in enabling cyberattacks, at 20%, while overall AI-related themes accounted for roughly 32% of all predictions.
The most common predictions weren’t about brand-new threats, but about familiar attacks becoming more efficient. AI was not expected to invent new attacks so much as to supercharge existing ones.
Examples of AI-driven attacks permeated headlines this year, including AI-assisted phishing attacks, automated reconnaissance, AI-optimized ransomware operations, and lower barriers to entry for attackers.
“As AI becomes more capable and accessible, the barrier to entry for less skilled attackers will become lower, while also accelerating the speed at which attacks can be carried out.”- Justin Blackburn, Senior Cloud Threat Detection Engineer, AppOmni
In fact, Google's threat intelligence research in November confirmed that by the end of 2025, threat actors were officially embracing AI-generated tools. like never before, to enhance all stages of the cyberattack lifecycle.
The first-of-its-kind AI malware was observed rewriting code "on the fly" to evade detection, while hacker marketplaces are now flooded with malicious, customizable AI tools at every price point.
Another trend predicted correctly was that SaaS and cloud risk would eclipse perimeter security as the primary battleground for preventing attacks.
“Many organizations lack visibility into their SaaS ecosystems, making it difficult to monitor user behavior, detect threats, and enforce security policies consistently.”- Ariel Parnes, Co-Founder & COO, Mitiga
Predictions clustered heavily around uncontrolled use of Software-as-a-Service (SaaS) applications, third-party integrations, API exposure, and identity and access misconfigurations, many of which led to multiple zero-day exploits. And let's not forget the 2025 cloud outages caused by misconfigurations, with AWS and Cloudflare outages causing worldwide havoc.
Bleeding into the ransomware category, third-party vendor ransomware attacks against large enterprises and major brands were rampant in 2025.
From the Cl0p gang’s latest Oracle e-Business Suite hacking campaign to the proliferation of Scattered Spider ransomware attacks on the retail sector worldwide, including the UK’s Marks & Spencer, ignoring the security of third-party tech providers is now obsolete.
Ransomware was viewed by many experts as a structural ecosystem problem, not a single threat to eliminate.
Experts further predicted that ransomware gangs would become more fragmented as law enforcement stepped up pressure, while competition, as well as cohesion, among groups and their members would also become more prevalent in 2025.
“The trend of ‘gang-hopping’ by individual cybercriminals between ransomware groups will further complicate attribution and containment efforts.”- Dr. Darren Williams, Founder & CEO, BlackFog
Again, right on the money, the lower entry for ransomware attackers has led to a 30-40% increase in the number of active groups, many of them smaller and harder-to-track operators, according to September research by Emsisoft.
Noteworthy predictions and where we go from here
Worth mentioning is regulation as a forecast topic. It made the top 10, but sadly not as a success story.
An increase in burdensome regulations and compliance appeared frequently among predictions, but largely became a complicating factor in 2025 rather than a solution.
Even before 2025 played out, experts had been skeptical that regulation would materially change the threat landscape, and they were right. More reporting requirements had little impact on attacker behavior.
“Increasing security risks and AI regulations on data handling will push organizations to enhance data visibility, classification, and governance.”- Bruno Kurtic, Co-Founder & CEO, Bedrock Security
The fifth top trend prediction was that data would overtake infrastructure as the primary security concern.
In fact, one could say that data has emerged as the core battleground underpinning nearly every major security concern, especially as data governance has become essential for AI use.
From data protection to 16 million password data leaks to the private user data being used to train every aspect of AI, data has surely become the ultimate buzzword of 2025, unless you count AI slop, of course.
If 2025 proved anything, it’s that the biggest cybersecurity risks weren’t the futuristic ones, they were the ones experts had been warning about all along.
As we head into 2026, the question isn’t whether these trends will continue, but whether organizations are any better prepared to confront them. We’ll see back here, same time next year, to find out.
Unlock more exclusive Cybernews content on YouTube.