Review: AI in Security & NLP on January 29, 2019

AI Security NPL Spike Reply Cluster Reply Logo 1


This article will soon be revised so that the impressions can be viewed correctly.

On January 29, 2019, members of the Microsoft AI User Group Cologne with Spike Reply in Cologne to talk about the Connections between artificial intelligence and security to discuss. Ibrahim Köse, Nils Milchert and Giovanni Battista Spinelli Barrile took the lead in highlighting the latest developments, possibilities and limitations of artificial intelligence and comparing them with traditional approaches and procedures. The evening was accompanied by delicious Pizza and drinks sponsored by Cluster Reply.


After an introductory welcome round with pizza and drinks, various topics were discussed:

  • Influence of artificial intelligence on proven security concepts and mechanisms
  • Added value of AI-supported security software (e.g. Windows Defender)
  • Weaknesses and manipulation possibilities of AI-based software (e.g. Twitter bot Tay)
  • Recognizing a mood image through natural language processing
  • Training sarcasm and irony

Particularly exciting was a Live demowhich showed how vulnerable today's production and manufacturing factories are to both modern and conventional malware. In particular, the lack of security and encryption protocols on machines, some of which are over 30 years old, is the basis for today's challenge of protecting one's own infrastructure in the best possible way.

How can you start creating your own AI applications? In addition to detailed help on Microsoft Learn Microsoft provides the Cognitive Services - a good starting point:

Take a look and discover More exciting meetups:


Leave a Reply

Your email address will not be published. Required fields are marked *