GNSEC Singapore Ltd - All are talking aboAut AI

but WE are talking about AI Security?!!

...

Artificial Intelligence - between Hype and Reality

It started a few years back...I remember we attended a security show. All products shown there were spiced up by AI and deep learning. And if you asked someone on the booth for more details, the answer was always the same: AI is making the product a better product, regardless how.

AI AND SECURITY

Talking to researchers about security and AI is in general pretty difficult. Most of them do not see any reasons for such a talk - they just play with data, so why should there be a security problem? Au contraire, security is the problem, because - from their view - it slows down their development. And well, Data is exactly one of the Problems, to be more specific, the integrity of the data. The machine learning process takes every input for granted, so the result is always based on the Data the system received - there is no input check. But ML can be fooled, Data can be manipulated. If there is no permanent control on the research results (output), bad things can happen.

AI is vulnerable

AI's can be vulnerable, can be manipulated or even taken hostage - by another, enemy AI. Or they can be misused, to do things, they were never meant to do. However, this is the same that can happen to traditional software - maybe not with this "humanlike" behaviour. Just remember input validation, so essential for any web-app. Without this, data can be misused or wrong interpreteted - and we can build a similar "misbehaviour" into an AI. But the developer can also prevent this - if he knows how and if he is sensitive for this topic.

AI Training

The most critical aspect of Ai and security is knowledge. The more all involved parties are educated on the topic, the better it is. We developed several individual training programs to help the customer to integrate AI and ML into their existing security strategy and projects. The training is addressing the different groups inside the organisation and is individually adapted to each customer.

Our Training Site

GNSEC Singapore Ltd - Our AI related Services

We offer AI related security consulting and training - inluding AI Threat Intelligence and Data Risk Modeling

Call us : +65 9111 1217 or send email to sales@gnesc.com for further inquiery.

    AI Security - Best Practises

  • Policymakers should collaborate closely with technical researchers to investigate, prevent, and mitigate potential malicious uses of AI.
  • Researchers and engineers in artificial intelligence should take the dual-use nature of their work seriously, allowing misuse- related considerations to influence research priorities and norms, and proactively reaching out to relevant actors when harmful applications are foreseeable.
  • Best practices should be identified in research areas with more mature methods for addressing dual-use concerns, such as computer security, and imported where applicable to the case of AI.
  • Actively seek to expand the range of stakeholders and domain experts involved in discussions of these challenges.

    • ...

    AI and Big Data Projects

    The provision of AI security is a very individual task and depends heavily on the project using AI or ML. Also, there is no framework or best practice paper - the topic is simply too new. That's why we use our experience as a long-standing security expert in software development and big data analysis.

    Our Services

  • Risk assessment / BIA of the project and the data used
  • Attack-Scenarios / Risk Analysis of the AI model and the ML Process.
  • General attack scenarios and vectors
  • Recommendations and QoS

    • Governance & Organization

    Big Data Analysis and Artificial Intelligence are not part of the corporate IT Security Framework. This is one reason, why it is so difficult to rule the projects. We cover he following topics with our Governance Support Program:

    Our Services

  • Security Controls for Big Data Analysis and AI
  • Best Practise Security Guidelines for AIDevelopment and ML.
  • AI related security organisation and skill requirements
  • Data Governance Security Models for ML

    • Training

    The most critical aspect of Ai and security is knowledge. The more all involved parties are educated on the topic, the better it is. We developed several individual training programs to help the customer to integrate AI and ML into their existing security strategy and projects. The training is addressing the different groups inside the organisation and is individually adapted to each customer:

    Our Services

  • Three to five days hybrid course on AI Security via GNSEC Academy (Q3/2023)
  • AI/ML related awareness and reality training
  • How to establish an AI Security Architekt?
  • Target Group oriented individual Security Training for AI/ML

    • AI - Today

    AI - The Future - maybe tomorrow....