AI News Roundup: AI Resurrects The Beatles for ‘Final’ Song
Also, IBM acquires a data company and unveils a coding assistant for watsonX
AI Business brings you the latest news and insights from across the AI world.
This week’s roundup covers the return of pop legends The Beatles thanks to AI, among other developments.
To keep up to date with coverage of all things AI, subscribe to the AI Business newsletter to get content straight to your inbox and follow the AI Business Podcast on Apple and Spotify.
The Beatles are back
The last Beatles song recorded with all four members was The End in 1969. But now, some 54 years on, the boys are back thanks to the power of AI.
A new song, titled ‘Now and Then’ will be published on Nov. 3. Lennon originally wrote the song and now half a century later, his voice will live on. To create Lennon’s vocals, Paul McCartney told BBC Radio 4 that AI was being used to “extricate” his voice from older recordings.
Upon hearing John’s voice again, McCartney said the experience was “quite emotional.”
“There it was, John’s voice, crystal clear. And we all play on it, it’s a genuine Beatles recording. In 2023 to still be working on Beatles music, and about to release a new song the public haven’t heard, I think it’s an exciting thing.”
Ringo Starr, the other surviving member of the Fab Four said: “It was the closest we’ll ever come to having him back in the room, so it was very emotional for all of us. It was like John was there, you know. It’s far out.”
The song will be released as a double-sided single along with ‘Love Me Do’ the Beatle’s debut single. Also being published is a documentary on its making, a day prior.
This is not the first time AI was used to bring the Beatles together again. Peter Jackson’s 2021 documentary ‘The Beatles: Get Back’ used AI to enhance the audio and images of footage of the band recording their final album.
IBM acquires data lineage company
IBM has acquired Manta Software, a data lineage platform to “complement its capabilities” with the company’s enterprise AI platform, watsonx.
Prague-based Manta was founded in 2016. Its platform is designed to give businesses improved visibility of their data environments. It provides an effective map of data flows, sources, transformations and dependencies to increase transparency for clients such as T-Mobile, BMO, J.B. Hunt and SCP Health.
It is now set to work within watsonx to improve data flow understanding for business users.
IBM had been working with Manta since June 2022 and already offer an integrated capability within IBM’s AI and data governance capabilities.
No financial details were disclosed.
IBM unveils watsonx Code Assistant
Separately, IBM launched a new code generation tool for watsonx.
Dubbed watsonx Code Assistant, the tool provides enterprise developers with the ability to generate code from natural language prompts.
The product covers two specific uses cases:
Watsonx Code Assistant for Red Hat Ansible Lightspeed - covers IT automation tasks like network configuration and code deployment.
Watsonx Code Assistant for Z - mainframe application modernization with the translation of COBOL to Java on IBM Z.1, first teased back in August.
The offerings are designed to be deployed in developer workflows “in a naturally integrated way that is designed to be non-disruptive – to help address skills gaps and increase productivity,” according to Kareem Yusuf, senior vice president of product management and growth at IBM Software.
Shutterstock launches new creative offerings
Stock image giant Shutterstock has launched new creative AI-powered editing features.
Now in beta, Shutterstock's creative AI editing features leverages tech from OpenAI to allow users to edit any image in its stock library.
Among the new tools is Magic Brush, which users can use to brush over areas they would like to change and describe what they want to be changed using natural language.
There is also the ability to generate alternate options of images with Variations, and Smart Resize, which automatically changes the shape of the image to match desired dimensions.
Shutterstock said artists will be compensated if their images are licensed after editing. However, users cannot publish AI-generated or edited images onto the Shutterstock platform for licensing.
"This new offering will bring our customers one step closer to their desired creative, as if they were directing the photoshoot themselves,” said Shutterstock CEO Paul Hennessy.
Telcos join forces to build large language models
Telco giants SK telecom and Deutsche Telekom have joined forces to develop large language models for the telecommunications sector.
The Korean and German companies, respectively, penned an agreement at a meeting in Seoul, South Korea. The jointly developed model will be built using technology from several AI companies, including Claude developer Anthropic and Facebook-parent Meta.
The model is set to be unveiled in the first quarter of 2024.
Read more about:
ChatGPT / Generative AIAbout the Author
You May Also Like