Meta, the parent company of Facebook and Instagram, has dissolved its Responsible AI (RAI) team in a surprising move and redirected its members to different departments within the company.
This move from the tech giant marks its significant shift in focus on spearheading generative AI. It raises questions about the commitment of Meta to ensure the responsible and ethical development of futuristic technologies.
Notably, Meta established its RAI team in 2019 to guide its technical teams on the ethical path while developing ML models. However, with the latest development, the Generative AI arm of Meta has absorbed most of these members.
This arm is tasked with the responsibility of generating content on the basis of prompts and requests from users. Besides, Meta has reassigned certain members to its ML infrastructure unit after disbanding the RAI team.
We continue to prioritize and invest in safe and responsible AI development, and these changes will allow us to better scale to meet our future needs.Meta spokesperson
The spokesperson added that most RAI members will work in the Generative AI wing, helping Meta work on responsible AI.
Meta’s decision to disband its RAI team has raised eyebrows about its commitment to developing responsible and ethical systems.
The CEO has also been vocal about his vision, the ‘Year of Efficiency’, where Meta’s focus is being redirected to generative AI.
Back in September 2022, Meta also dissolved its Responsible Innovation team, which used to work under its Social Impact department. The latest move brings the question of its ability to maintain an effective ethical oversight.
Interestingly, Meta fired thousands of employees earlier this year.
It’s evident that Zuckerberg is streamlining technical roles in the organization, as Meta discontinued low-priority projects like the custom silicon unit in its Reality Labs and the Fundamental AI Research department with its AI protein-folding team.
Zuckerberg said, “Our single largest investment is in advancing AI and building it into every one of our products”. The CEO also stated that they already have the necessary infrastructure in place to realize these goals rapidly.
Meta Continues To Launch Generative AI Products
In recent months, Meta has launched a plethora of generative AI products. These include AI-generated stickers, Llama large language models, Emu for creating animated images, and ML services for advertising.
They have also developed Emu Video and Emu Edit to allow users to control the model through written commands and convert text to video. Meta aims to use these tools across its apps on social media platforms where users can edit videos and photos or even develop customized GIFs in their posts and messages.
As the social media giant shifts its focus to internal restructuring in an effort to remain at the forefront, the dissolution of the Responsible AI team marks its progressive stance on the innovative front. It remains to be seen how Meta manages to ensure the ethical development of its AI tools with its RAI team dissolved.