Scanning and analyzing a blockchain can be difficult, especially when there are tens of thousands of blockchain addresses and millions of transactions to filter through. It’s laborious and time-consuming, and the tale is frequently elusive. Blocktrace, an Austin-based startup, thinks that artificial intelligence could speed up the blockchain analysis process and make detecting trends and anomalies easier.
Blocktrace is an AI-powered blockchain forensic and analytical service provider founded in 2018 by software engineer Shaun MaGruder, who formerly worked as the head of training at blockchain forensics startup Chainalysis. Blocktrace used an AI to interact with data on the Bitcoin blockchain by using a chatbot called Robby the Robot, named after the classic character from the sci-fi film “Forbidden Planet.”
Blocktrace CEO MaGruder told Decrypt in an interview that the business chose Bitcoin for the Robby project due of its vast sample size and history.
“Bitcoin is a terrific initiative to start with and, like Ethereum, will most likely always be around,” MaGruder remarked. “It’s regarded as the original because it was the first and has a vast pool of user base addresses with a high level of activity.”
According to MaGruder, a copy of the Bitcoin blockchain data is already saved in Blocktrace’s database, which is an impressive technical feat. Their program, which has been enhanced with an AI layer, allows users to ask natural language inquiries like a virtual assistant.
A natural language model is a sort of artificial intelligence that understands and processes human language in the same way that humans do.
“Users can immediately request specific information, such as transactions between dates and amounts,” stated MaGruder. “As compared to the manual method, which could take one to two hours using a blockchain explorer, this saves time.”
According to MaGruder, the goal of Blocktrace is to allow investigators and users to use OpenAI’s technology to swiftly detect Bitcoin addresses and identify transactions that occurred on the network with more accuracy and detail.
“Instead of requiring a data engineer or data scientist to translate natural language questions into SQL queries, Robbie has already been trained on the data model and can quickly fetch results for the user,” MaGruder explained, calling the process more efficient and eliminating the need for new data scientists to undergo continuous training.
While currently in development, Robby is scheduled to be released to the public later this year. However, before releasing Robby to the general public, MaGruder stated that the chatbot would first be tested on a small group of people.
“We want to make sure Robbie is steady and that we don’t blow it when we release it to the public,” MaGruder explained. “We want to make sure that the public’s questions are questions that we’ve already asked Robbie.”
“As expected,” MaGruder stated, “people will compare Robbie to a regular blockchain explorer to assess whether Robbie is correct or erroneous.” “To collect input, we will implement a thumbs up and thumbs down button, with a thumbs up serving as positive reinforcement and a thumbs down indicating areas in need of improvement.”
Blocktrace is the most recent Web3 business to seek to harness the potential of artificial intelligence for blockchain analysis. Elliptic, Chainalysis, CipherTrace, and Nansen are among the companies that use the technology.
“It’s something we’re looking into really carefully,” Andrew Thurman, a Nansen developer at ETH Denver, told Decrypt. “You can use things like AI analysis on the blockchain to check if [wallets] are doing the same things at the same time,” Thurman explained, adding that even if no explicit connections exist, you can infer with a high degree of assurance that all of those wallets are linked to the same entity.
“I think algorithmic and AI labeling have a really substantial future in blockchain analysis,” Thurman added.
