Tolley.Ai applies machine learning to tax
Tax data and training provider Tolley is the latest organisation to embrace machine learning with Tolley.Ai, a query-handling engine for tax topics.
Now available in an experimental version for public testing, Tolley.Ai draws on a collection of self-contained data sets relating to basic tax definitions, rates and allowances, tax calculations.
The best way to test its capabilities is to visit the Tolley.Ai website and test it with a question relating to those basic topics. A query about the married couples allowance, for example, successfully differentiated it from the newer marriage allowance, and flagged up the current rates that apply, the conditions the claimants need to fulfil (one of them born before 6 April 1935, living together and resident in the EU etc).
Tolley's head of tax workflow Onwah Tsang, who is leading the artificial intelligence project, played down expectations. This is not going to be tax equivalent of Deep Thought from ‘Hitchhiker’s Guide to the Galaxy’ that comes up with the ultimate tax answers and make judgments on the merits and legality of a particular treatment.
Instead, Tolley.Ai is a conversational system designed to interpret the underlying intention behind tax questions and select the most suitable result from the existing database.
“It’s similar to the way people interact with Google – which is good at interpreting natural language,” Tsang said.
As Tolley.Ai learns from these initial encounters and proves its worth, the publisher will bring feed other databases into the mix to extend its capabilities.
“It could act as an augumentation tool where a skilled professional can get guidance. It helps alleviate risk, because it will help them consider every angle,” Tsang said.
To explore the emerging possibilities of AI, open banking and digital tax, Tolley is organising a Future of Tax Technology meeting in London on the evening of 24 April. Information about the event is provided by a version of Tolley.Ai.
You might also be interested in
AccountingWEB’s interim Editor in Chief has been with the site since 1999 and returned to the editorial hot seat in March 2020 to lead the hunt for a long-term successor... Send a DM if you're interested! When not tending to the needs of AccountingWEB members and geeking out on their technology habits, he devotes much of his time to his oddball...