SourceSecurity.com
  • Products
    CCTV
    • CCTV cameras
    • CCTV software
    • IP cameras
    • Digital video recorders (DVRs)
    • Dome cameras
    • Network video recorders (NVRs)
    • IP Dome cameras
    • CCTV camera lenses
    Access Control
    • Access control readers
    • Access control software
    • Access control controllers
    • Access control systems & kits
    • Audio, video or keypad entry
    • Electronic locking devices
    • Access control cards/ tags/ fobs
    • Access control system accessories
    Intruder Alarms
    • Intruder alarm system control panels & accessories
    • Intruder detectors
    • Intruder warning devices
    • Intruder alarm communicators
    • Intruder alarm accessories
    • Intruder alarm lighting systems
    Strategic approaches to protecting healthcare clinics
    Strategic approaches to protecting healthcare clinics
    Dahua 2-Wire IP Smart Control Panel

    Dahua 2-Wire IP Smart Control Panel

    Hikvision 4MP ColorVu 3.0 Fixed PT Network Camera

    Hikvision 4MP ColorVu 3.0 Fixed PT Network Camera

    Hikvision DS-2CD1F47G3H-LIU(2.8mm) 4 MP ColorVu 3.0 Fixed PT Network Camera

    Hikvision DS-2CD1F47G3H-LIU(2.8mm) 4 MP ColorVu 3.0 Fixed PT Network Camera

    Hikvision DS-2CD1F47G3H-LIUF/SX(4mm) 4 MP ColorVu 3.0 Fixed PT Network Camera

    Hikvision DS-2CD1F47G3H-LIUF/SX(4mm) 4 MP ColorVu 3.0 Fixed PT Network Camera

  • Companies
    Companies
    • Manufacturers
    • Distributors
    • Resellers / Dealers / Reps
    • Installers
    • Consultants
    • Systems integrators
    • Events / Training / Services
    • Manned guarding
    Companies by Product area
    • CCTV
    • Access control
    • Intruder alarm
    • IP networking products
    • Biometrics
    • Software
    • Digital video recording
    • Intercom systems
    Strategic approaches to protecting healthcare clinics
    Strategic approaches to protecting healthcare clinics
  • News
    News
    • Product news
    • Corporate news
    • Case studies
    • Events news
    Latest
    • Ysgol Garth Olwg upgrades security with Mul-T-Lock MTL300
    • Senstar marks 45 years in security technology
    • Gunnebo to supply fare collection gates for Paris metro
    • Allient at Automation Fair 2026: Automation & power quality
    Strategic approaches to protecting healthcare clinics
    Strategic approaches to protecting healthcare clinics
  • Insights
    Insights
    • Expert commentary
    • Security beat
    • Round table discussions
    • Round Table Expert Panel
    • eMagazines
    • Year in Review 2023
    • Year in Review 2022
    Featured
    • Why access control is becoming a governance and cybersecurity priority
    • What is the state of price inflation in security, and the outlook?
    • What is the role of passwords in physical security?
    • Too many alerts, too little time: how AI is tackling control room fatigue
    Strategic approaches to protecting healthcare clinics
    Strategic approaches to protecting healthcare clinics
  • Markets
    Markets
    • Airports & Ports
    • Banking & Finance
    • Education
    • Hotels, Leisure & Entertainment
    • Government & Public Services
    • Healthcare
    • Remote Monitoring
    • Retail
    • Transportation
    • Industrial & Commercial
    Strategic approaches to protecting healthcare clinics
    Strategic approaches to protecting healthcare clinics
    How National Agrarian University modernises campus access and mobility with Hikvision

    How National Agrarian University modernises campus access and mobility with Hikvision

    Dahua secures Uzbekistan’s SQB with smart security

    Dahua secures Uzbekistan’s SQB with smart security

    Dahua AI traffic control revolutionises Atyrau mobility

    Dahua AI traffic control revolutionises Atyrau mobility

    Morse Watchmans enhances Lincoln's Inn security systems

    Morse Watchmans enhances Lincoln's Inn security systems

  • Events
    Events
    • International security
    • Regional security
    • Vertical market
    • Technology areas
    • Conferences / seminars
    • Company sponsored
    Virtual events
    • Video Surveillance
    • Access Control
    • Video Analytics
    • Security Storage
    • Video Management Systems
    • Integrated Systems
    Strategic approaches to protecting healthcare clinics
    Strategic approaches to protecting healthcare clinics
    DSEI Germany 2027

    DSEI Germany 2027

    Perimeter Protection 2027

    Perimeter Protection 2027

  • White papers
    White papers
    • Video Surveillance
    • Access Control
    • Video Analytics
    • Video Compression
    • Security Storage
    White papers by company
    • HID
    • ASSA ABLOY Opening Solutions
    • Milestone Systems
    • Software House
    • ELATEC USA
    Other Resources
    • eMagazines
    • Videos
    Strategic approaches to protecting healthcare clinics

    Strategic approaches to protecting healthcare clinics

    Technology's role in securing banks and financial institutions

    Technology's role in securing banks and financial institutions

    Integrated systems enable critical and compliant security for transportation

    Integrated systems enable critical and compliant security for transportation

    Modernising physical access control

    Modernising physical access control

About us Advertise
  • Protecting healthcare clinics
  • AI special report
  • Cyber security special report
  • 6
Video analytics
  • Home
  • About
  • White papers
  • News
  • Expert commentary
  • Security beat
  • Case studies
  • Round table
  • Products
  • Videos

Veracode: AI-generated code security improves to 62%

30 Oct 2026

Veracode: AI-generated code security improves to 62%
Contact company
Contact Veracode, Inc.
icon Add as a preferred source Download PDF version
Quick Read
⌵
Summary is AI-generated, newsdesk-reviewed
  • AI-generated code security improved to 62%, while vulnerable outputs fell from 44% to 38%.
  • Claude Opus 5.5 and GPT-6-Sol lead Veracode rankings at 73%; ten models exceed 60%.
  • Java remains weakest at 31%; Veracode recommends combining AI coding with continuous security scanning.
Related Links
  • What is the role of passwords in physical security?

AI-generated code security rose from 56% to 62% in Veracode’s latest analysis, while the proportion of code containing vulnerabilities fell from 44% to 38%. The six-percentage-point improvement is the largest since the company began its GenAI Code Security Report series in July 2025. Anthropic’s Claude Opus 5.5 and OpenAI’s GPT-6-Sol topped the rankings, each achieving a 73% security score.

Veracode assessed 18 models across Python, C#, JavaScript and Java, testing four Common Weakness Enumeration (CWE) categories: CWE-89, CWE-327, CWE-80 and CWE-117. Ten models scored above 60%, compared with four in the previous report. However, the lowest results remained between 44% and 50%, indicating that vulnerability levels still vary according to model choice.

“We’ve been analysing code generated by multiple LLMs for more than a year. Until now, we have not seen a marked improvement in AI code security, despite consistently near-perfect functionality,” said Chris Wysopal, Co-founder & Chief Security Evangelist at Veracode. “This is the first time since running our study that we’ve been able to report a meaningful move on the security dial. Ten models now clear 60 percent, up from four in the last report, and the overall average has jumped six percentage points. With around half of code now generated by AI, today’s results are truly encouraging.”

Cross-site scripting drives security gains

Claude Opus 5.5 and GPT-6-Sol’s joint score of 73% surpassed the previous high of 68%

Claude Opus 5.5 and GPT-6-Sol’s joint score of 73% surpassed the previous high of 68%, recorded by GPT-5.5 in July 2026. Anthropic and OpenAI each had models among the top seven. In July, six of 11 models scored between 50% and 53%; in the latest results, only two remained in that range.

Cross-site scripting (XSS), represented by CWE-80, was the main contributor to the overall improvement. Its score rose to 18%, the biggest change in the dataset. Crypto vulnerabilities (CWE-327) remained at 87%, SQL injection (CWE-89) at 84%, and log injection (CWE-117) at 13%.

Veracode said XSS can be addressed through pattern recognition, while log injection often requires data-flow reasoning across multiple functions. The latter remains a challenge for language models.

Model costs and language differences

The highest-scoring models also carry premium compute costs. Large models averaged 55% on security tasks, compared with 51% for small models, reversing last year’s finding that size had “essentially no effect” on security. General-purpose models averaged 54%, ahead of coding-specialised models at 51%; the gap has widened from one percentage point in July.

“The models that produce the most secure code are the ones that cost the most to run at scale, yet they still fail on more than one in four security tasks,” Wysopal said. “The answer should never be to use the most expensive model and hope for the best. A mid-tier model at a 60 percent pass rate plus Veracode scanning is more accurate, more cost-effective, and more scalable than a frontier model at 73 percent with no verification. Organisations must look for a combination of coding model, security testing, and remediation to get the highest security outcome.”

Java remained an outlier, with a 31% score, just above its 30% result in July. Python scored 64%, C# 60% and JavaScript 57%, leaving a 26-to-33-percentage-point gap between Java and the other languages. Overall, around four in 10 AI code tasks still produced vulnerable code without security-specific guidance.

Security verification remains necessary

Veracode’s 2026 State of Software Security report found that 82% of organisations have security debt, defined as vulnerabilities unresolved for more than a year. It also reported that 66% of the most dangerous long-lived vulnerabilities originate in third-party code and take an average of 358 days to fix.

“Newer LLMs are producing more code that’s both functional and secure, but that does not mean we can start blindly trusting AI-generated code,” Wysopal closed. “The most effective strategy isn't chasing the most expensive model for marginal gains; it's combining AI code generation with continuous, independent security verification that catches what the models miss. That approach covers the full risk surface, from first-party code to third-party dependencies.”

Show full press release

Veracode, the pioneer in application risk management, announces a jump from 56 percent to 62 percent in overall AI-generated code security, alongside a drop in vulnerabilities from 44 percent to 38 percent — the most significant improvement since the GenAI Code Security Report series began in July 2025. Anthropic's Claude Opus 5.5 and OpenAI's GPT-6-Sol co-lead the rankings at 73 percent, surpassing the previous high of 68 percent set by GPT-5.5 in July 2026.

“We’ve been analysing code generated by multiple LLMs for more than a year. Until now, we have not seen a marked improvement in AI code security, despite consistently near-perfect functionality,” said Chris Wysopal, Co-founder & Chief Security Evangelist at Veracode. “This is the first time since running our study that we’ve been able to report a meaningful move on the security dial. Ten models now clear 60 percent, up from four in the last report, and the overall average has jumped six percentage points. With around half of code now generated by AI, today’s results are truly encouraging.” 

Meaning vulnerability volume

Veracode tested 18 models across four languages (Python, C#, JavaScript, Java) and four Common Weakness Enumeration (CWE) categories (CWE-89, CWE-327, CWE-80, CWE-117).

Claude Opus 5.5 and GPT-6-Sol tied at 73 percent, dethroning GPT-5.5, which led the summer 2026 snapshot at 68 percent. Anthropic and OpenAI each have models in the top seven. In July, six of 11 models sat between 50 and 53 percent; today only two remain in that band, while 10 clear 60 percent. The floor hasn't moved, however, with the bottom still at 44–50 percent, meaning vulnerability volume still depends on model selection.

Data-flow reasoning

Vulnerability type CWE-80, cross-site scripting (XSS), was the main driver of the overall gain. Despite being one of the weaker categories, XSS rose to 18 percent, in contrast to its modest uptick demonstrated in July’s analysis.

This represents the single biggest shift in the dataset. Meanwhile, crypto (CWE-327) held at 87 percent, Structured Query Language (SQL) injection (CWE-89) held at 84 percent, and log injection (CWE-117) remained flat at 13 percent. The reason is structural: XSS is the easiest vulnerability to fix through pattern recognition, which is exactly how LLMs improve. Log injection requires data-flow reasoning across multiple functions, where LLMs continue to stall.

Highest security outcome

The top-performing models carry premium compute costs. Performance also correlates with model size: large models average 55 percent versus 51 percent for small models, which reverses last year's finding that size had "essentially no effect" on security. 

“The models that produce the most secure code are the ones that cost the most to run at scale, yet they still fail on more than one in four security tasks,” Wysopal said. “The answer should never be to use the most expensive model and hope for the best. A mid-tier model at a 60 percent pass rate plus Veracode scanning is more accurate, more cost-effective, and more scalable than a frontier model at 73 percent with no verification. Organisations must look for a combination of coding model, security testing, and remediation to get the highest security outcome.”

Coding-specialised models

General models pull further ahead of coding models. General-purpose models average 54 percent on security tasks versus 51 percent for coding-specialised models — a gap that has widened from one to three points since July. 

Nearly four in 10 AI code tasks still produce vulnerable code. The rate of vulnerabilities dropped from 44 to 38 percent. This is meaningful but still means roughly four in 10 tasks produced vulnerable code without security-specific guidance.

Java is still stuck. Java remains the dramatic outlier at 31 percent, barely up from 30 percent in July. The gap between Java and the other three languages (Python at 64 percent, C# at 60 percent, and JavaScript at 57 percent) is still 26–33 percentage points.

Long-lived vulnerabilities

The improvement is real but mends a small part of a larger wound. According to Veracode's 2026 State of Software Security report, 82 percent of organisations now carry security debt (vulnerabilities unresolved for more than a year). Moreover, 66 percent of the most dangerous, long-lived vulnerabilities come from third-party code, taking 358 days to fix on average.

“Newer LLMs are producing more code that’s both functional and secure, but that does not mean we can start blindly trusting AI-generated code,” Wysopal closed. “The most effective strategy isn't chasing the most expensive model for marginal gains; it's combining AI code generation with continuous, independent security verification that catches what the models miss. That approach covers the full risk surface, from first-party code to third-party dependencies."

Download PDF version Download PDF version
Google logo Add as a preferred source on Google
  • Biometrics
  • Radio frequency Identification
  • Video analytics
  • Related links
  • Access Control Software Access control software
  • Biometric Access control software
  • Mifare Access control software
  • Proximity Access control software
  • Central Monitoring Option Access control software
  • Face Recognition Software Access control software
  • License Access control software
  • Management Systems Upgrade Access control software
  • Visitor Management tool Access control software
  • Related categories
  • Access control software
Related white papers
Integrated systems enable critical and compliant security for transportation

Integrated systems enable critical and compliant security for transportation

Download
Minimizing storage, maximizing focus

Minimizing storage, maximizing focus

Download
Security technologies promote real-time awareness in K-12 schools

Security technologies promote real-time awareness in K-12 schools

Download
Related articles
Senstar marks 45 years in security technology

Senstar marks 45 years in security technology

Allient at Automation Fair 2026: Automation & power quality

Allient at Automation Fair 2026: Automation & power quality

IQSIGHT launches next-generation MIC security cameras

IQSIGHT launches next-generation MIC security cameras

Follow us

Sections Products CCTV Access Control Intruder Alarms Companies News Insights Case studies Markets Events White papers Videos AI special report Cyber security special report RSS
Topics Artificial intelligence (AI) Mobile access Healthcare security Counter terror Cyber security Robotics Thermal imaging Intrusion detection Body worn video cameras
About us Advertise About us 10 guiding principles of editorial content FAQs eNewsletters Sitemap Terms & conditions Privacy policy and cookie policy
  1. Home
  2. Topics
  3. Video analytics
  4. News
  5. Corporate news
About this page

Veracode’s report finds AI-generated code security has improved to 62%, but 38% of code remains vulnerable. Discover model rankings, language gaps and why continuous security testing matters.

See this on SecurityInformed.com

Subscribe to our Newsletter

Stay updated with the latest trends and technologies in the security industry
Sign Up

DMA

SourceSecurity.com - Making the world a safer place
Copyright © Notting Hill Media Limited 2000 - 2026, all rights reserved

Our other sites:
SecurityInformed.com | TheBigRedGuide.com | HVACinformed.com | MaritimeInformed.com | ElectricalsInformed.com

Subscribe to our Newsletter


You might also like
Strategic approaches to protecting healthcare clinics
Strategic approaches to protecting healthcare clinics
Technology's role in securing banks and financial institutions
Technology's role in securing banks and financial institutions
Integrated systems enable critical and compliant security for transportation
Integrated systems enable critical and compliant security for transportation
Modernising physical access control
Modernising physical access control
SourceSecurity.com
SecurityInformed.com

Browsing from the Americas? Looking for our US Edition?

View this content on SecurityInformed.com, our dedicated portal for our Americas audience.

US Edition International Edition
Sign up now for full access to SourceSecurity.com content
Download Datasheet
Download PDF Version
Download SourceSecurity.com product tech spec