Open Access

Das Projekt PLA2Scale entwickelt nachhaltige, recyclingfähige Lebensmittelverpackungen aus PLA-Blends. Ziel ist es, biobasierte Kunststoffe wie PLA wirtschaftlich und ökologisch sinnvoll im Verpackungsbereich zu etablieren – durch Materialoptimierung, Tests, Recycling und ökologische Bewertung.

Die neue EU-Verpackungsverordnung und ihre Chancen und Herausforderungen.

Aufgrund des hohen Energieverbrauchs und des Einsatzes zahlreicher Single-Use-Materialien rückt das Thema Nachhaltigkeit in der Reinraumtechnik zunehmend in den Fokus. Diese Entwicklung wird auch durch politische Vorgaben bestärkt. Unter Berücksichtigung der Balance zwischen einer Reduktion des CO2-Fußabdrucks und der Wahrung strenger Reinheitsstandards ergeben sich vielversprechende Lösungsansätze insbesondere durch die Steigerung der Energieeffizienz im Betrieb und den Einsatz nachhaltiger Verbrauchsmaterialien.

This article investigates the potential misuse of large language models (LLM) for low-resource, highly personalised social engineering attacks. The study explores how ChatGPT can infer personality traits during natural conversations by leveraging publicly available personal information, such as social media data, as an entry point. Utilising the social engineering personality framework (SEPF), the research endeavours to optimise attack vectors based on the Big Five personality traits, with the objective of enhancing the persuasiveness of social engineering strategies. The approach is divided into four phases: verifying conversational capabilities, conducting personality analyses, applying the SEPF for attack optimisation, and evaluating the persuasiveness of personalised attacks. The present paper offers a proof-of-principle for the initial phase, demonstrating ChatGPT’s capacity to engage in natural conversations while conducting personality analyses in a discreet manner. The findings indicate that while ChatGPT exhibits the capacity to simulate human-like interactions, limitations in conversational variance and the reliability of personality assessment were observed. The study identifies challenges such as generalisations, lack of score differentiation, and confirmation bias, and proposes refinements like increasing interaction depth, adjusting scoring scales, and using tailored personas. Subsequent research will investigate enhanced personality inference techniques, personalisation of attack vectors, and their impact on susceptibility to social engineering attacks.

Although laboratory automation has been used successfully in some areas for decades, the general knowledge about laboratory automation and its user-friendliness still leaves a lot to be desired in some areas of application. In my presentation, I would like to introduce our open-source and easily adaptable voice-user-interface system “Rainbow“ for laboratory environments. “Rainbow“ achieves a speech recognition accuracy of at least 98.6%. On the other hand, I would like to present our decision-making aid on how interested laboratories can introduce laboratory automation systems and thereby are able to avoid common mistakes.Finally, I would like to talk about how we are training the next generations of laboratory automation specialists at our university.