Google Wants to Humanize Artificial Intelligent

Google recently launched PAIR, an acronym of People + AI Research, in an attempt to increase the utility of AI and improve human to AI interaction. The initiative unites different Google researchers to find new ways to use AI, rather than using AI to better their current services.

Currently, companies like Google and Facebook use AI for identifying people and objects in pictures or to quickly translate text on web pages. However, there’s a shortage of new services and apps based on AI for direct use by consumers. This is what the PAIR initiate hopes to research and develop.

“We don’t have all the answers—that’s what makes this interesting research—but we have some ideas about where to look,” wrote Martin Wattenberg and Fernanda Viégas, two researchers on the project.

PAIR has a variety of goals to guide the research. For example, they would like to discover uses of AI for professionals, like doctors, farmers, designers, musicians, and so on that they can use everyday in their work, separate from the other Google products currently offered, though as of yet, there are not specifics for what exactly what this will entail.

Additionally, PAIR would like to increase the accessibility to machine learning AI breakthroughs by attracting lead developers in the field. However, critics, especially in academia, say more people would have access if big companies like Google weren’t hoarding AI developers for themselves.

Finally, PAIR will also focus on troubleshooting guidelines and tools for developers of AI as traditional debugging techniques do not work as well on AI, which is constantly changing due to its machine learning.

More News to Read

Also published on Medium.