Skip to main content

Brain transplant: Watson AI to make Slack’s chatbot smarter

Introducing Watson Virtual Agent
IBM’s Watson supercomputing AI, which previously beat human competitors at Jeopardy and even edited its own magazine, is set to impart some of its wisdom to another automated system: Slack’s chatbot. It’s all part of IBM’s new Watson Virtual Agent effort, which allows Watson’s intelligence to be imparted to all sorts of human interface scenarios.

While the idea of having conversations with artificial intelligence might seem like a gateway to a Skynet future, chatbots are seen as a massive growth industry in the coming years. Employed by the likes of Facebook, health care institutions, and even public service websites. Watson is ahead of the curve in this area, so applying its level of conversational ability to something like Slack makes a lot of sense.

As it stands, Slack’s helper bot is rather simplistic. It can answer basic questions about how to use Slack, and can allow you to set reminders or searches through the help system for you to find relevant information. It’s handy, but hardly versatile. The idea behind the Watson implementation is to broadly increase its functionality.

Related: Watson makes its debut in the U.S. Open mobile app, making tennis fans smarter than ever

It will become more capable of answering complex queries, and will enable the chatbot’s communications to appear more lifelike and conversational (thanks Engadget).

This is just the start of Watson’s global rollout. After the Slackbot has been upgraded with its new intelligence suite, Slack and IBM will share their findings with other developers, making it possible for anyone to use the Watson Conversation API. That way, any number of developers can create new and intelligent chatbots.

IBM is already working on a new chatbot that could act as a network administrator of sorts, answering questions outages, network integrity, and more. This should take some of the load off of real network administrators, and also provide better and faster information to those who might need it.

Jon Martindale
Jon Martindale is the Evergreen Coordinator for Computing, overseeing a team of writers addressing all the latest how to…
A dangerous new jailbreak for AI chatbots was just discovered
the side of a Microsoft building

Microsoft has released more details about a troubling new generative AI jailbreak technique it has discovered, called "Skeleton Key." Using this prompt injection method, malicious users can effectively bypass a chatbot's safety guardrails, the security features that keeps ChatGPT from going full Taye.

Skeleton Key is an example of a prompt injection or prompt engineering attack. It's a multi-turn strategy designed to essentially convince an AI model to ignore its ingrained safety guardrails, "[causing] the system to violate its operators’ policies, make decisions unduly influenced by a user, or execute malicious instructions," Mark Russinovich, CTO of Microsoft Azure, wrote in the announcement.

Read more