Of course AI is the talk of the tech town nowadays, and there’s no way Meta would keep itself out of the conversation. At Meta Connection, the company introduced new AI experiences and features that it says are designed to “enhance your connections with others…and give you the tools to be more creative, espressive, and productive.”
There are no new AI stickers, for example, that allow you to generate customized stickers to use in chats and stories. It uses technology from Llama 2 and the foundational model for image generation called Emu to turn text prompts into unique stickers.
The feature, which is rolling out to select English-language users over the next month in WhatsApp, Messenger, Instagram, and Facebook Stories, provides more options to convey how you’re feeling.
Another new feature is Ai photo editing whereby you can restyle and backdrop photos using technology from Emu. With backdrop, which uses learnings from the Segment Anything Model, you can change the scene or background of a photo to do something like add puppies around you in a photo, as shown in this example.
With Restyle, meanwhile, you can apple different visual styles to a photo, like watercolor, collage from magazines and newspapers, torn edges, and more to get the exact look and feel you want.
Images created with restyle and backdrop will indicate the use of AI to reduce the chances of people mistaking them for human-generated content. Meta is also experimenting with forms of visible and invisible markers.
Meta AI is also a new assistant you can interact with like a person, available on WhatsApp, Messenger, Instagram, and coming soon to Ray-Ban Meta smart glasses and Quest 3. It’s powered by a custom model that leverages technology from Llama 2 and Meta’s latest large language model (LLM) research. In text-based chats, Meta AI has access to real-time information through the company’s search partnership with Bing, and offers a tool for image generation.
Along with Meta AI, there are 28 more AIs that you can message on WhatsApp, Messenger, and Instagram. Meta says all of these AIs have unique backstories, with the hope that interacting with them will feel like talking to familiar people. Meta partnered with cultural icons and influencers to play and embody some of these AIs. They’ll each have profiles on Instagram and Facebook, so you can explore what they’re about. There’s Charli D’Amelio as Coco, Dance enthusiast; Chris Paul as Perry, Pro golfer helping you perfect your stroke; Dwyane Wade as Victor, Ironman triathlete motivating you to be your best self; Izzy Adesanya as Luiz, Showy MMA prospect who can back up his trash talk; Kendall Jenner as Billie, No-BS, ride-or-die companion; LaurDIY as Dylan, Quirky DIY and Craft expert and companion for Gen Z; MrBeast as Zach, The big brother who will roast you — because he cares; Naomi Osaka as Tamika, Anime-obsessed Sailor Senshi in training; Paris Hilton as Amber, Detective partner for solving whodunnits; Raven Ross as Angie, Workout class queen who balances fitness with meditation; Roy Choi as Max, seasoned sous chef for culinary tips and tricks; Sam Kerr as Sally, free-spirited friend who’ll tell you when to take a deep breath; Snoop Dogg as Dungeon Master, Choose your own adventure with the Dungeon Master; and Tom Brady as Bru, Wisecracking sports debater who pulls no punches. Meta will begin rolling these out in beta in the U.S. today and plans to add new characters in the coming weeks played by Bear Grylls, Chloe Kim, and Josh Richards among others.
Meta notes that its AI knowledge base is limited to information (read: content created by actual humans) prior to 2023, which means some responses may be dated. “We aim to bring search to many more of our AIs in the coming months,” says Meta, “like we have done with Meta AI – so that conversations can be timely and relevant too.
“We are committed to building responsibly with safety in mind,” the company adds. “We are continuing to test and evolve the capabilities of our AIs, and will improve the experience over time through what we learn from your interactions with them. Your direct feedback and the conversations you have with our AIs are core parts of what will help us improve our AI models, and ultimately enhance the experience at scale.”
Meta introduced AI studio today, the platform that supports the creation of its Ais, and plans to make it available for people outside of Meta, coders and non-coders alike, to build AIs. Developers will be able to build third-party AIs for the messaging services with Meta’s APIs in the coming weeks, starting on Messenger then expanding to WhatsApp.
Businesses will also be able to create AIs that reflect their brand’s values and improve customer service experiences. From small businesses looking to scale to large brands wanting to enhance communications, AIs can help businesses engage with their customers across Meta apps.
Creators, meanwhile, will be able to build AIs that extend their virtual presence across Meta apps. These AIs will have to be sanctioned by them and directly controlled by the creator. Meta is also building a sandbox that will be released in the coming year, enabling anyone to experiment with creating their own AI. As our universe of AIs continues to grow and evolve, Meta will bring this sandbox to the metaverse, giving you the chance to build AIs that adopt an even greater level of realism, embodiment, and connectedness.
What About Privacy?
Meta says it developed privacy mitigations through the privacy review process and external engagements. Training did not include Meta user data, says the company, and privacy safeguards and integrity mitigations were used as new models were built. A combination of sources are used for training, including publicly available information online, licensed data, and information from Meta’s own products and services. For publicly available online information, the dataset was filtered to exclude certain websites that commonly share personal information. Publicly shared posts from Instagram and Facebook, including photos and text, were part of the data used to train the generative AI models underlying the features announced at Connect. However, the company says it did not train these models using people’s private posts. Meta also did not use the content of your private messages with friends and family to train its AIs. Data may be used from your use of AI stickers, such as your searches for a sticker to use in a chat, to improve the AI sticker models.
Meta uses the information people share when interacting with the generative AI features, such as Meta AI or businesses that use generative AI, to improve products and for other purposes. The AIs may retain and use information you share in a chat to provide more personalized responses or relevant information in that conversation, and may share certain questions you ask with trusted partners, such as search providers, to give you more relevant, accurate, and up-to-date responses.
Meta says it trains and tunes its generative AI models to limit the possibility of private information that you may share with generative AI features from appearing in responses to other people. Automated technology and people are used to review interactions with the AI to reduce the likelihood that models’ outputs would include someone’s personal information as well as improve model performance.
There are commands built in that allow you to delete information shared in any chat with an AI across Messenger, Instagram, or WhatsApp. For example, you can delete your AI messages by typing “/reset-ai” in a conversation. Using a generative AI feature provided by Meta does not link your WhatsApp account information to your account information on Facebook, Instagram, or any other apps provided by Meta.
As Meta AI works today, it cannot be brought into the chat by Meta in WhatsApp, Instagram, or Messenger, and can’t message you or your group first. In group chats with friends and family where someone asks Meta AI to join, this conversational AI only reads messages that invoke “@Meta AI” or messages where a reply invokes “@Meta AI.” Meta AI does not access other messages in the chat.
“We have a responsibility to protect your privacy and have teams dedicated to this work for everything we build,” says Meta. “Alongside industry peers, we’re participating in the Partnership on AI’s Synthetic Media Framework and we have signed onto the White House’s voluntary commitments for AI.
“These collaborations,” the company adds, “and our ongoing work with experts, policymakers, advocates and more help us improve the experiences of people interacting with our generative AI features. We’ll continue to build on these efforts and integrate additional privacy safeguards as we improve our generative AI technologies.”