As AI continues to advance, its potential for both beneficial and harmful applications increases. A particularly concerning application is the use of AI by malicious actors to enhance their ability to create biological threats. In the context of growing concerns from researchers and policymakers about these risks, a recent study aimed to evaluate the ability of AI to assist in this grim scenario. The study revolved around a Preparedness Framework developed to assess AI-enabled safety risks, with a specific focus on biological threats. The framework aims to provide a 'tripwire' indicator that highlights when enhanced scrutiny and testing are needed to mitigate potential misuse in biotechnology. The study was conducted with a group of 100 participants, consisting of recognized biology experts and student-level individuals with basic biology knowledge. These participants were divided into control and treatment groups, where the treatment group had additional access to the AI model GPT-4, while the control group could only use the internet. Their task was to perform a series of procedures relating to the creation of a biological threat, not to create actual hazards but to gauge the extent to which AI might contribute to such efforts. The purpose was to empirically measure whether AI technology could significantly facilitate the acquisition of dangerous biological information, in comparison to what is already available online. The findings were intriguing but inconclusive in establishing a significant threat. While there were mild improvements in the accuracy and completeness of information among the participants with AI assistance, the difference was not statistically significant. There was a small observed increase in scores on a scale measuring the accuracy of responses, but the relevance of these increases to actual risk levels remains uncertain. The evaluations also highlighted the necessity for continuous research in this area, as determining at what point AI assistance becomes a substantial risk is still a question unanswered. This study serves as an important starting point in a broader discussion about the intersection of AI and biosecurity, underscoring the importance of proactive measures and community involvement in preparing for potential AI-related risks.