Microsoft dials again Bing after customers handle to recreate Disney emblem in faux AI-generated photographs
AI In Transient Microsoft is updating the Bing AI chatbot service to stop customers producing faux movie posters containing Disney’s emblem over fears of copyright infringement.
Folks on social media started sharing their AI-generated creations of canines as Disney characters mimicking its movie posters on-line. The pattern could have caught the eye of the leisure behemoth, main Microsoft to quickly block the phrase “Disney” in folks’s prompts, in line with the Monetary Instances.
The ban has since been lifted, however Bing reportedly would not recreate Disney’s wording or emblem anymore. The incident has revitalized arguments that AI text-to-image fashions violate copyright since they weren’t solely skilled on protected IP or logos, however faithfully generate the content material too.
This is an example of the well-known, iconic “Afghan woman” picture from Nationwide Geographic being reproduced by Midjourney.
Artists and illustrators suing corporations constructing AI text-to-image instruments, like Stability and Midjourney, have argued that the software program can straight rip off their work by copying their types. The defendants, nonetheless, consider that AI-generated photographs must be protected below honest use because the photographs remodel authentic works and don’t exchange them.
However whether or not they really are transformative sufficient is an open query, particularly if they’ll recreate the photographs they had been skilled on.
Amazon lays off a whole lot of staff engaged on Alexa
Amazon has axed a whole lot of workers on the Alexa group because it shifts its focus to generative AI, in line with an inner memo.
Daniel Rausch, VP of Alexa and Hearth TV, introduced the choice this week, confirming “a number of hundred roles” had been going to be “eradicated.”
“We’re actually inspired by the progress we’re making with Alexa: clients are interacting with Alexa tens of tens of millions of instances each hour and there are greater than half a billion Alexa units in clients’ properties. Our investments in generative AI are additionally bringing our imaginative and prescient for an much more intuitive, clever, and helpful Alexa nearer than ever earlier than,” the word, first reported by GeekWire, started.
“As we proceed to invent, we’re shifting a few of our efforts to raised align with our enterprise priorities, and what we all know issues most to clients — which incorporates maximizing our sources and efforts centered on generative AI. These shifts are main us to discontinue some initiatives, which is leading to a number of hundred roles being eradicated.”
It is not clear what these roles entail precisely, and the way this can change Alexa as a product.
Rausch mentioned that the corporate will proceed to concentrate on attempting to include a brand new massive language mannequin to energy the voice-activated AI assistant.
Airbnb acquires secret stealth AI startup
Airbnb has snapped up a secretive AI startup named GamePlanner.AI a deal price $200 million to reportedly develop the final word “journey concierge”.
Little is thought about GamePlanner.AI, or what expertise it has been constructing to draw Airbnb’s consideration. Based in 2020 by Adam Cheyer and Siamak Hodjat, the 12-person group has operated in stealth mode.
Cheyer and Hodjat are greatest identified for constructing voice-activated chatbots. Cheyer has a protracted historical past within the business and cofounded Siri, the place Hodjat was a senior engineer, earlier than Apple purchased it in 2010. The pair moved to Apple earlier than leaving to start out Viv Labs, one other startup that developed software program that may later grow to be the Bixby chatbot after being acquired by Samsung.
Airbnb’s CEO Brian Chesky beforehand mentioned he needs to make use of generative AI to construct a journey concierge able to studying about customers’ preferences over time so it may possibly discover the perfect properties for them to remain in, in line with CNBC. The house-rental market biz gave little away in its assertion saying the deal, solely stating that GamePlanner.AI “will concentrate on accelerating choose AI tasks and integrating their instruments.”
Cheyer will be part of Airbnb as VP of AI Expertise, and Hodjat shall be a director. “A giant a part of what attracted us to Airbnb was our shared dedication to utilizing AI to allow human connection,” Cheyer mentioned. “Like Brian, I consider that with out nice design and community-based intelligence, AI can solely obtain a fraction of its potential.”
Bard for teenagers
Google is opening up its Bard web chatbot for youngsters to make use of.
In most nations, these 13 and older which have a Google account will have the ability to discuss to Bard. Youngsters in Europe, Switzerland, and the UK, nonetheless, should be 18 or over. In South America, they should be at the least 14 and over, and 15 in Korea. The net app, nonetheless, will solely be accessible in English for these below 18.
Google needs teenagers to have the ability to work together with Bard to study issues like methods to brainstorm concepts or ask questions on particular subjects. They’ll additionally get Bard to assist them study abilities or full homework. Given a picture of a math equation, Bard can, for instance, clarify the issue and provides hints on methods to clear up it, Google claimed.
There are extra safeguards in place when youngsters use the software program. Bard ought to keep away from speaking about inappropriate issues that may be doubtlessly dangerous like unlawful medication. Its responses will routinely embody a “G” icon youthful customers can click on on to go looking Google, a characteristic the corporate calls “double-check response.”
“We acknowledge that many individuals, together with teenagers, should not all the time conscious of hallucinations in LLMs. So the primary time a teen asks a fact-based query, we’ll routinely run our double-check response characteristic, which helps consider whether or not there’s content material throughout the net to substantiate Bard’s response,” Google mentioned in a weblog submit. ®
