Analysis of AI Tools Shows 85 Percent Have Been Breached

AI tools are becoming essential to modern work, but their fast, unmonitored adoption is creating a new kind of security risk. Recent surveys reveal a clear trend – employees are rapidly adopting consumer-facing AI tools without employer approval, IT oversight, or any clear security policies. According to Cybernews Business Digital Index, nearly 90% of analyzed AI tools have been exposed to data breaches, putting businesses at severe risk.

About 75% of workers use AI in the workplace, with AI chatbots being the most common tools to complete work-related tasks. While this boosts productivity, it could expose companies to credential theft, data leaks, and infrastructure vulnerabilities, especially since only 14% of workplaces have official AI policies, contributing to untracked AI use by employees.

While a significant number of employees use AI tools at work, a large share of this usage remains untracked or unofficial. Estimates show that around one-third of AI users keep their usage hidden from management.

Personal accounts are used uncontrollably for work tasks

According to Google’s 2024 survey of over 1,000 U.S.-based knowledge workers, 93% of Gen Z employees aged 22–27 use two or more AI tools at work. Millennials aren't far behind, with 79% reporting similar usage patterns. These tools are used to draft emails, take meeting notes, and bridge communication gaps.

Additionally, a 2025 Elon University survey found that 58% of AI users regularly rely on two or more different models, while data from Harmonic indicates that 45.4% of sensitive data prompts are submitted using personal accounts, completely bypassing company monitoring systems.

“Unregulated use of multiple AI tools in the workplace, especially through personal accounts, creates serious blind spots in corporate security. Each tool becomes a potential exit point for sensitive data, outside the scope of IT governance,” says Emanuelis Norbutas, Chief Technical Officer at nexos.ai, a secure AI orchestration platform for businesses. “Without clear oversight, enforcing policies, monitoring usage, and ensuring compliance becomes nearly impossible.”

Most popular AI tools struggle with cybersecurity

To better understand how these tools perform behind the scenes, Cybernews researchers analyzed 52 of the most popular AI web tools in February 2025, ranked by total monthly website visits based on Semrush traffic data.

Using only publicly available information, Business Digital Index uses custom scans, IoT search engines, IP, and domain name reputation databases to assess companies based on online security protocols.

The findings paint a concerning picture. Widely used AI platforms and tools show uneven and often poor cybersecurity performance. Researchers found major gaps despite an average cybersecurity score of 85 out of 100. While 33% of platforms earned an A rating, 41% received a D or even an F, revealing a deep divide between the best and worst performers.

“What is mostly concerning is the false sense of security many users and businesses may have,” says Vincentas Baubonis, Head of Security Research at Cybernews. “High average scores don’t mean tools are entirely safe – one weak link in your workflow can become the attacker’s entry point. Once inside, a threat actor can move laterally through systems, exfiltrate sensitive company data, access customer information, or even deploy ransomware, causing operational and reputational damage.”

84% of AI tools analyzed have suffered data breaches

Out of the 52 AI tools analyzed, 84% had experienced at least one data breach. Data breaches often result from persistent weaknesses like poor infrastructure management, unpatched systems, and weak user permissions. However, even more alarming is that 36% of analyzed tools experienced a breach in just the past 30 days.

Alongside breaches, 93% of platforms showed issues with SSL/TLS configurations, which are critical for encrypting communication between users and tools. Misconfigured SSL/TLS encryption weakens the protection of data sent between users and platforms, making it easier for attackers to intercept or manipulate sensitive information.

System hosting vulnerabilities were another widespread concern, with 91% of platforms exhibiting flaws in their infrastructure management. These issues are often linked to weak cloud configurations or outdated server setups that expand the attack surface.

Password reuse and credential theft

44% of companies developing AI tools showed signs of employee password reuse – a significant enabler of credential-stuffing attacks, where hackers exploit recycled login details to access systems undetected.

In total, 51% of analyzed tools have had corporate credentials stolen, reinforcing the need for stronger password policies and IT oversight, especially as AI tools become routine in the workplace. Credential theft is often a forerunner to a data breach, as stolen credentials can be used to access sensitive data.

“Many AI tools simply aren’t built with enterprise-grade security in mind. Employees often assume these tools are safe by default, yet many have already been compromised, with corporate credentials among the first targets,” says Norbutas. “When passwords are reused or stored insecurely, it gives attackers a direct line into company systems. Businesses must treat every AI integration as a potential entry point and secure it accordingly.”

Productivity tools show weakest cybersecurity

Productivity tools, commonly used for note-taking, scheduling, content generation, and work-related collaboration, emerged as the most vulnerable category, with vulnerabilities across all key technical domains. Particularly infrastructure, data handling, and web security.

According to Business Digital Index analysis, this category had the highest average number of stolen corporate credentials per company (1,332), and 92% had experienced a data breach. Every single tool in the category had 100% system hosting and SSL/TLS configuration issues.

“This is a classic Achilles’ heel scenario,” says cybersecurity expert Baubonis. “A tool might appear secure on the surface, but a single overlooked vulnerability can jeopardize everything. Hugging Face is a perfect example of that risk – it only takes one blind spot to undermine months of security planning and expose the organization to threats it never anticipated.”

Research Methodology

Cybernews researchers examined 52 of the 60 most popular AI tools in February 2025, ranked by total monthly website visits based on Semrush traffic data. Seven tools could not be scanned due to domain limitations.

The report evaluates cybersecurity risk across seven key dimensions: software patching, web application security, email protection, system reputation, hosting infrastructure, SSL/TLS configuration, and data breach history.

Featured

  • Allegion, Comfort Technologies Implement Mobile Credentials at the Artisan Apartment Homes in Florida

    Artisan Apartment Homes, a luxury apartment complex in Dunedin, Florida, recently transitioned from mechanical keys to electronic locks and centralized system software with support from Allegion US, a leading provider of security solutions, technology and services, and Florida-based Comfort Technologies, which specializes in deploying multifamily access control, IoT devices and software management solutions. Read Now

  • Mall of America Deploys AI-Powered Analytics to Enhance Parking Intelligence

    Mall of America®, the largest shopping and entertainment complex in North America, announced an expansion of its ongoing partnership with Axis Communications to deploy cutting-edge car-counting video analytics across more than a dozen locations. With this expansion, Mall of America (MOA) has boosted operational efficiency, improved safety and security, and enabled more informed decision-making around employee scheduling and streamlining transportation for large events. Read Now

  • Security Industry Association Launches New “askSIA” AI Tool

    The Security Industry Association (SIA) has unveiled a brand-new SIA member benefit – askSIA, a conversational AI agent designed to help users get the most out of their SIA membership, easily access SIA resources and find the latest information on SIA’s training and courses, reports and publications, events, certification offerings and more. SIA members can easily find askSIA by visiting the SIA homepage or looking for the askSIA icon in the top left of webpages. Read Now

    • Industry Events
  • Industry Embraces Mobile Access, Biometrics and AI

    A combination of evolving workplace dynamics, technology innovation and new user expectations is changing how people enter and interact with physical spaces. Access control is at the heart of these changes. Combined with biometrics and AI, mobile access control has become increasingly crucial for deploying entry solutions that are seamless, secure and adaptive to user needs. Read Now

  • Sustainable Video Solution Delivered for Landmark City of London Office Development

    An advanced, end-to-end video solution from IDIS, with a focus on reducing waste and costs, has helped a major office development in the City of London align its security with sustainability objectives. Read Now

New Products

  • Connect ONE’s powerful cloud-hosted management platform provides the means to tailor lockdowns and emergency mass notifications throughout a facility – while simultaneously alerting occupants to hazards or next steps, like evacuation.

    Connect ONE®

    Connect ONE’s powerful cloud-hosted management platform provides the means to tailor lockdowns and emergency mass notifications throughout a facility – while simultaneously alerting occupants to hazards or next steps, like evacuation.

  • Camden CM-221 Series Switches

    Camden CM-221 Series Switches

    Camden Door Controls is pleased to announce that, in response to soaring customer demand, it has expanded its range of ValueWave™ no-touch switches to include a narrow (slimline) version with manual override. This override button is designed to provide additional assurance that the request to exit switch will open a door, even if the no-touch sensor fails to operate. This new slimline switch also features a heavy gauge stainless steel faceplate, a red/green illuminated light ring, and is IP65 rated, making it ideal for indoor or outdoor use as part of an automatic door or access control system. ValueWave™ no-touch switches are designed for easy installation and trouble-free service in high traffic applications. In addition to this narrow version, the CM-221 & CM-222 Series switches are available in a range of other models with single and double gang heavy-gauge stainless steel faceplates and include illuminated light rings.

  • PE80 Series

    PE80 Series by SARGENT / ED4000/PED5000 Series by Corbin Russwin

    ASSA ABLOY, a global leader in access solutions, has announced the launch of two next generation exit devices from long-standing leaders in the premium exit device market: the PE80 Series by SARGENT and the PED4000/PED5000 Series by Corbin Russwin. These new exit devices boast industry-first features that are specifically designed to provide enhanced safety, security and convenience, setting new standards for exit solutions. The SARGENT PE80 and Corbin Russwin PED4000/PED5000 Series exit devices are engineered to meet the ever-evolving needs of modern buildings. Featuring the high strength, security and durability that ASSA ABLOY is known for, the new exit devices deliver several innovative, industry-first features in addition to elegant design finishes for every opening.