Safeguarding LLMs against Misuse and AI-Driven Malware Using Steganographic Canaries

AI-powered malware increasingly exploits cloud-hosted generative-AI services and large language models LLMs as analysis engines for reconnaissance and code generation. Simultaneously, enterprise uploads expose sensitive documents to third-party AI vendors. Both threats converge at the AI service...

Design Principles for the Construction of a Benchmark Evaluating Security Operation Capabilities of Multi-Agent AI Systems

As Large Language Models LLMs and multi-agent AI systems are demonstrating increasing potential in cybersecurity operations, organizations, policymakers, model providers, and researchers in the AI and cybersecurity communities are interested in quantifying the capabilities of such AI systems to...

CVE-2026-4997

A security flaw has been discovered in Sinaptik AI PandasAI up to 3.0.0. This affects the function issqlquerysafe of the file pandasai/helpers/sqlsanitizer.py. Performing a manipulation results in path traversal. The attack may be initiated remotely. The exploit has been released to the public an...

CVE-2026-33879

Federated Learning and Interoperability Platform FLIP is an open-source platform for federated training and evaluation of medical imaging AI models across healthcare institutions. The FLIP login page in versions 0.1.1 and prior has no rate limiting or CAPTCHA, enabling brute-force and...

CVE-2026-4965

A vulnerability was detected in letta-ai letta 0.16.4. This issue affects the function resolvetype of the file letta/functions/astparsers.py of the component Incomplete Fix CVE-2025-6101. Performing a manipulation results in improper neutralization of directives in dynamically evaluated code. The...

EUVD-2026-16925

A weakness has been identified in Sinaptik AI PandasAI up to 3.0.0. This vulnerability affects the function CodeExecutor.execute of the file pandasai/core/codeexecution/codeexecutor.py of the component Chat Message Handler. Executing a manipulation can lead to code injection. The attack may be...

CVE-2026-4997

A security flaw has been discovered in Sinaptik AI PandasAI up to 3.0.0. This affects the function issqlquerysafe of the file pandasai/helpers/sqlsanitizer.py. Performing a manipulation results in path traversal. The attack may be initiated remotely. The exploit has been released to the public an...

CVE-2026-4998 Sinaptik AI PandasAI Chat Message code_executor.py CodeExecutor.execute code injection

A weakness has been identified in Sinaptik AI PandasAI up to 3.0.0. This vulnerability affects the function CodeExecutor.execute of the file pandasai/core/codeexecution/codeexecutor.py of the component Chat Message Handler. Executing a manipulation can lead to code injection. The attack may be...

EUVD-2026-16921

A vulnerability was identified in Sinaptik AI PandasAI up to 0.1.4. Affected by this issue is the function deletequestionandanswers/deletedocs/updatequestionanswer/updatedocs/getrelevantquestionanswersbyid/getrelevantdocsbyid of the file extensions/ee/vectorstores/lancedb/pandasailancedb/lancedb....

CVE-2026-4997 Sinaptik AI PandasAI sql_sanitizer.py is_sql_query_safe path traversal

A security flaw has been discovered in Sinaptik AI PandasAI up to 3.0.0. This affects the function issqlquerysafe of the file pandasai/helpers/sqlsanitizer.py. Performing a manipulation results in path traversal. The attack may be initiated remotely. The exploit has been released to the public an...

CVE-2026-4997

A security flaw has been discovered in Sinaptik AI PandasAI up to 3.0.0. This affects the function issqlquerysafe of the file pandasai/helpers/sqlsanitizer.py. Performing a manipulation results in path traversal. The attack may be initiated remotely. The exploit has been released to the public an...

CVE-2026-22743

Spring AI's spring-ai-neo4j-store contains a Cypher injection vulnerability in Neo4jVectorFilterExpressionConverter. When a user-controlled string is passed as a filter expression key in Neo4jVectorFilterExpressionConverter of spring-ai-neo4j-store, doKey embeds the key into a backtick-delimited...

CVE-2026-22742

Spring AI's spring-ai-bedrock-converse contains a Server-Side Request Forgery SSRF vulnerability in BedrockProxyChatModel when processing multimodal messages that include user-supplied media URLs. Insufficient validation of those URLs allows an attacker to induce the server to issue HTTP requests...

CVE-2026-22744

In RedisFilterExpressionConverter of spring-ai-redis-store, when a user-controlled string is passed as a filter value for a TAG field, stringValue inserts the value directly into the @field:VALUE RediSearch TAG block without escaping characters.This issue affects Spring AI: from 1.0.0 before 1.0....

WordPress AI Engine (Pro) plugin < 3.4.2 - Broken Access Control vulnerability

Broken Access Control vulnerability discovered by Trương Hữu Phúc truonghuuphuc in WordPress Plugin AI Engine Pro versions 3.4.2...

PT-2026-28715

Name of the Vulnerable Software and Affected Versions Sinaptik AI PandasAI versions up to 3.0.0 Description A security flaw exists in Sinaptik AI PandasAI up to version 3.0.0. The issue resides within the is sql query safe function located in the pandasai/helpers/sql sanitizer.py file, allowing f...

PT-2026-28714

Name of the Vulnerable Software and Affected Versions Sinaptik AI PandasAI versions up to 0.1.4 Description A SQL injection issue exists in the pandasai-lancedb Extension within Sinaptik AI PandasAI. The issue is located in the file extensions/ee/vectorstores/lancedb/pandasai lancedb/lancedb.py a...

EUVD-2026-16736

A vulnerability was detected in letta-ai letta 0.16.4. This issue affects the function resolvetype of the file letta/functions/astparsers.py of the component Incomplete Fix CVE-2025-6101. Performing a manipulation results in improper neutralization of directives in dynamically evaluated code. The...

CVE-2026-4965

A vulnerability was detected in letta-ai letta 0.16.4. This issue affects the function resolvetype of the file letta/functions/astparsers.py of the component Incomplete Fix CVE-2025-6101. Performing a manipulation results in improper neutralization of directives in dynamically evaluated code. The...

CVE-2026-4965

CVE-2026-4965 affects letta-ai letta 0.16.4. The vulnerability resides in letta/functions/ast_parsers.py, in the resolve_type function, where improper neutralization of directives in dynamically evaluated code enables remote arbitrary-code execution. The issue is linked to an incomplete fix for C...