🕵️‍♀️ AI takes the stand

What’s the fastest, most reliable witness at a crime scene? No, it’s not a camera, it’s a fly. For decades, the tiny, timed life cycle of the blowfly (which includes, wait for it, maggots) has been a standard tool for estimating a corpse’s time of death, a precise but painfully slow process — until AI stepped in. You can read more about that in Meghan Rosen’s fascinating SN article on how scientists are using an AI technique to identify different varieties of these corpse-eating insects, to better fight crimes.

🪰 Like clockwork … with wings

The blowfly life cycle acts like a precise stopwatch, with different larval (read: maggot) stages linked directly to time. Historically, forensic scientists had to manually collect larvae and figure out which species they came from to link to that species’ exact stopwatch. Figuring that out is not only painstaking and highly sensitive to human error but also requires specialized entomological expertise. A team led by Rabi Musah, an organic chemist at Louisiana State University in Baton Rouge, is leveraging deep learning models trained on chemical fingerprints from hundreds of the cocoon-esque casings left behind after those maggots grow their wings. Investigators could analyze casings to get a precise determination of the species, which could help them determine time of death.

⚖️ The case for scale

Specialized forensics suffers from a critical lack of scalability in manual and biological protocols. Expert forensic entomologists are scarce. One traditional species identification technique, rearing any larvae found on a corpse to adulthood (because eggs and larvae from many species tend to look alike) can take days or weeks, delaying case resolution. AI bypasses this bottleneck; Musah’s technique shrinks turnaround time from days to about 90 seconds. This democratization of expertise might also reduce the costs of complex crime scene analysis, allowing valuable human experts to focus on nuanced case interpretation rather than basic data collection.

🪲 The VC swarm

While funding for specialized entomology AI is still in its infancy, here are a few other applications of tech-enhanced forensics.

• The US federal government supports a network of Regional Computer Forensics Laboratories. Many university forensics labs (like the one mentioned in the blowfly article) secure grants from agencies like the National Institute of Justice to develop AI-based classification systems. They apply deep learning to problems like fiber analysis, paint chip matching or toolmark identification (determining whether a specific tool was used to leave a mark at a crime scene). The laboratories often rely on commercial forensic software Cellebrite (NASDAQ:CLBT), which went public in 2018 via a $2.4 billion SPAC deal.
  
• Verogen is a leading commercial platform for forensic DNA analysis. The company was acquired by Netherlands-based QIAGEN in 2023 for $150 million.
  
• Magnet Forensics develops software for forensic analysis of computers, cloud services and mobile devices. Their tools use machine learning to identify known child exploitation images and automatically flag relevant data points across massive datasets. This Canadian company was acquired in 2023 by American private equity firm Thoma Bravo for $1.8 billion.

The takeaway? Maggots or not, AI is sure to be a key factor in future crime fighting.

---

Disclaimer: This newsletter is for informational purposes only and does not constitute investment advice. Society for Science and Science News Media Group assumes no liability for any financial decisions or losses resulting from the use of the content in this newsletter. Society for Science and Science News Media Group do not receive payments from, and do not have any ownership or investment interest in, the companies mentioned in this newsletter. Please consult a qualified financial advisor before making any investment decisions.