Do Generative AI, Augmented Reality, and Virtual Reality shape tomorrow’s B2B sales? “Yes!” say industry colleagues and experts on Tuesday at the fourth DIGITAL COMMERCE SUMMIT automotive & industry in Saarbrücken Castle. Even at the pre-event, including dinner speeches and delicious tapas, new contacts were made, and old ones were refreshed. On the event day… Continue reading Do Generative AI, Augmented Reality, and Virtual Reality shape tomorrow’s B2B sales?
Merry Christmas and a Happy New Year 2024! Once again, the end of the year is knocking at the door. The past 12 months have been an exciting journey filled with thrilling and successful projects, and, above all, marked by our collaborative partnership. This collaboration and your trust have made our year truly special, and… Continue reading Merry Christmas and a Happy New Year 2024!
BSI ISSUES WARNING ABOUT INDIRECT PROMPT INJECTION Application-oriented language models exhibit threats due to intrinsic vulnerabilities. In February 2023, we discovered this in collaboration with CISPA – Helmholtz Center for Information Security. Now, the Federal Office for Information Security (BSI) has also published a cybersecurity warning regarding the security gap we found, cautioning against the… Continue reading BSI ISSUES WARNING ABOUT INDIRECT PROMPT INJECTION
SIMPLY EXPLAINED – INDIRECT PROMPT INJECTION The concept of Indirect Prompt Injection may appear relatively simple at first glance, potentially leading to the misconception that it is harmless. However, it can quickly become quite complex. To truly grasp its depth, listening to the interview with our employee Kai on the MLSecOPs Podcast is worthwhile. For… Continue reading SIMPLY EXPLAINED – INDIRECT PROMPT INJECTION
HITTING THE HEADLINES IN SAARLAND Our research on Indirect Prompt Injections in Large Language Models is making headlines internationally and locally. We are delighted about the coverage in the Saarbrücker Zeitung ( https://www.saarbruecker-zeitung.de/saarland/saarland-informatiker-finden-sicherheitsluecke-bei-chatgpt_aid-88337939 ) and the Aktueller Bericht of Saarländischer Rundfunk (https://www.sr-mediathek.de/index.php?seite=7&id=125471).
SEQUIRE-INTERVIEW IN ZEIT-ONLINE MAGAZIN One of the first to delve into the Indirect Prompt-Injection vulnerability we discovered was the award-winning journalist Eva Wolfangel, who works for Zeit Online. We are thrilled about the comprehensive report; collaborating with her was a pleasure. For the full article, click here: https://www.zeit.de/digital/2023-03/cyberangriffe-microsoft-bing-chat-piraten/komplettansicht
SEQUIRE GOES INTERNATIONAL In the international press, VICE Motherboard was one of the first portals to publish a comprehensive report on our research. For the complete coverage, follow the link below: https://www.vice.com/en/article/7kxzzz/hackers-bing-ai-scammer