You are viewing the chat in desktop mode. Click here to switch to mobile view.
X
AI-driven stories: How journalist can work with robots and algorithms
powered byJotCast
Pratik Aswalkar
3:30
In Gen Summit, 2019, Lisa Gibbs, director of news partnerships at The Associate Press, Ben Rudolph, Managing Director of Microsoft News Labs, will discuss how journalists will work with AI technology to transform journalism with moderator Reginald Chua, COO of Reuters.
3:38
Ben Rudolph leads Microsoft News Labs, Microsoft’s global effort to help journalists and journalism succeed by augmenting human creativity with innovative AI, collaboration, and content creation technology
3:40
Lisa Gibbs is the director of news partnerships at The Associated Press and the Newsroom lead on AP’s automation and artificial intelligence strategy group.
3:41
Reginald Chua manages editorial operations, budgets, administration, safety, and security for a 2,500-person newsroom; oversees data journalism and graphics; and liaises with technology and R&D teams to build newsroom capabilities and tools at Reuters.
3:45
Reginald Chua begins the discussion by explaining the lack of information about capabilities of Artificial Intelligence in the public domain.
3:47
Chua (to Lisa Gibbs): "How is your organization struggling, if at all its struggling, to keep up with AI. How is it affecting the news creation in the newsroom?"
3:49
Gibbs responds, "From the perspectives of the associated press, using AI was all about efficiency. As the AI has gotten more experienced, the challenges have become more sophisticated."
3:51
Gibbs explains how journalists will work with robots and how AI can be used in journalists' workflow, especially in areas of news gathering.
These are prime areas of exploration, she says, in which AI could play a big role.
3:53
Chua asks Rudolph to share his experience the creation and process of AI and role of AI in journalistic experiences.
3:56
Rudolph responds, "The biggest shift AI has caused is in how story is created and also the presentation of things have changed '"steep hockeysticking of democratization tools".
Will AI completely replace journalists in the future?
Yes (0% | 0 votes)
 
No (0% | 0 votes)
 
Total Votes: 0
3:59
Rudolph adds, "Earlier you would need multiple types of specialists. So only an organisation like AP, Reuters etc which had the people and machinery to hire so many people could do  big digitial stories but now as technological reach has increased, more people can use digital tools to make impactful videos."
4:02
Ben Rudolph on AI
4:03
Chua chimes in with an example about how got 183 flight passes and helicopters are hovering and reporting instantly.
Rudolph summarises by explaining how it is much easier to train an AI program.
Gibbs adds, "AI tools can help us create different versions of stories for multiple platforms at the same time."
4:04
He continues, "One of the things AP is working on is a summarization tool. The robot can produce a summary that's meant to be spoken of every AP text."
4:06
Chua poses the question to Gibbs, "What is a journalists' role there when a machine is producing more than 50 innovative stories. How do you innovate in the narratives and facts?"
4:07
Lisa quips, "Even if the content is automatically generated, it won't be automatically published. It changes the relationship between the editor and the reporter now that there's an additional tool for fact-finding."
4:09
Chua to Rudolph: "How close are we to being able to create a truly interactive experience instead of a pre-recorded AI?"
4:10
Rudolph: "It depends on the breadth of experience you are trying to deliver. Conversations can be really complicated and are extraordinarily large amounts of data is created.
So instead of focusing on all, we will focus on depth-based experiences amongst narrower topic ranges."
4:11
4:12
Rudolph continues, "For example, Liverpool is my favourite team - getting data like how many goals, who scored etc is structured data - easily available.
What happened in the world of sports, today is a complex question. It's a broad question. What team I like in different sports - a boxing man will ask this question, but the AI will know that he meant boxing updates and not general sports. This is a more natural way of interacting."
4:15
Chua: "How hard is collecting data in sports in an age of machinated minute details?"
4:16
Gibbs inquires on how Chua would answer the question of journalists doing stories in an AI world.
4:18
Chua replies, "the idea obviously is to do multiple versions of the stories." He elaborates on interactive chatbot, which changes the whole definition of a journalists' job where it puts forward facts instead of storytelling.
4:20
Rudolph interjects, "When I start to think about what I need to do as an individual is to create an AI experience. Instead of having tools to feed inputs and create an AI experience, it creates a more natural and interactive experience."
4:21
Chua counter-questions Rudolph, "why keep it in the newsroom instead of selling it as a product?"
4:22
-Reginald Chua
4:24
Rudolph adds, "with reference to news organizations, some of the world's best reporting is happening for the last 200 years in the basements in London. For example, we know this information is there, but we can't reach it.
In this context, content created by AI can be used as an additional experience provided by big news houses on subscription bases."
4:27
Gibbs says that she's working a lot on AI assistance to journalists.
4:28
Chua replies to GIbbs, "what he wants is nicely summarised story and not a 50 various glimpses of a story."
4:30
Gibbs explains that there is still a long way to go to help be the reporter more insightful. So AI is not just a fact gatherer.
"We are currently rolling out through most of AP event detection tools. When we tested this tool with 27 journos in AP, we never used the word AI. It's just like any other technology in the world, it has it's implications.
"We didn't say Oh Hey! Now you're using AI in your job "
4:31
Rudolph gives his final comment : people are becoming extremely sceptical."
They question everything.
We (Microsoft team) want to create a tool that optimises the creative skills of our people and give the best output for the viewers.
4:33
Before the Q-A session, Chua ends the discussion by saying, "It is also true that what you optimise the data for, is also very critical."
Connecting…