Webinar "Machine Learning Security: Natural Problems of Artificial Intelligence"

On March 21, Digital Security experts will present their critical look at the widespread adoption of artificial intelligence (AI) and machine learning (MO).

No matter how ideal the capabilities of AI are idealized today, this is an algorithm that can make mistakes, and not a checkmate to alternative methods of solving the problem. We will discuss the specific vulnerabilities of the algorithms, identify the resulting reputation and financial risks and ways to increase the stability of the algorithms.

In the webinar program:

• AI and MO in the modern world: areas of use, the range of tasks;
• Real cases of error algorithms and the problems that they entail;
• Adversarial attacks on algorithms;
• Ways to increase the stability of algorithms, protection methods.

The webinar will be useful:

• Company executives, IT directors and development directors who review and implement AI technologies;
• Technical specialists working with AI and MO algorithms;
• Representatives of a wide audience interested in AI issues.

The webinar will be held March 21 at 11:00 (Moscow time). Mandatory registration .