News: 1700491511

  ARM Give a man a fire and he's warm for a day, but set fire to him and he's warm for the rest of his life (Terry Pratchett, Jingo)

Microsoft dials back Bing after users manage to recreate Disney logo in fake AI-generated images

(2023/11/20)


AI In Brief Microsoft is updating the Bing AI chatbot service to prevent users generating fake film posters containing Disney's logo over fears of copyright infringement.

People on social media began sharing their AI-generated creations of dogs as Disney characters mimicking its film posters online. The trend may have caught the attention of the entertainment behemoth, leading Microsoft to temporarily block the word "Disney" in people's prompts, according to the Financial Times.

The ban has since been lifted, but Bing reportedly doesn't recreate Disney's wording or logo anymore. The incident has revitalized arguments that AI text-to-image models violate copyright since they were not only trained on protected IP or trademarks, but faithfully generate the content too.

[1]

Here's [2]an example of the famous, iconic "Afghan girl" photo from National Geographic being reproduced by Midjourney.

[3]

[4]

Artists and illustrators [5]suing companies building AI text-to-image tools, like Stability and Midjourney, have argued that the software can directly rip off their work by copying their styles. The defendants, however, believe that AI-generated images should be protected under fair use since the images transform original works and do not replace them.

But whether they truly are transformative enough is an open question, especially if they can recreate the images they were trained on.

Amazon lays off hundreds of employees working on Alexa

Amazon has axed hundreds of staff on the Alexa team as it shifts its focus to generative AI, according to an internal memo.

Daniel Rausch, VP of Alexa and Fire TV, announced the decision this week, confirming "several hundred roles" were going to be "eliminated."

[6]

"We're really encouraged by the progress we're making with Alexa: customers are interacting with Alexa tens of millions of times every hour and there are more than half a billion Alexa devices in customers' homes. Our investments in generative AI are also bringing our vision for an even more intuitive, intelligent, and useful Alexa closer than ever before," the note, [7]first reported by GeekWire, began.

"As we continue to invent, we're shifting some of our efforts to better align with our business priorities, and what we know matters most to customers — which includes maximizing our resources and efforts focused on generative AI. These shifts are leading us to discontinue some initiatives, which is resulting in several hundred roles being eliminated."

It's not clear what those roles entail exactly, and how this will change Alexa as a product.

[8]

Rausch said that the company will continue to focus on trying to incorporate a new large language model to power the voice-activated AI assistant.

[9]AI copyright row deepens: Stability VP quits in protest over 'fair use' excuse

[10]To pay or not to pay for AI's creative 'borrowing' – that is the question

[11]OpenAI hits the GPT-4 Turbo button plus promises copyright shield for fans

Airbnb acquires secret stealth AI startup

Airbnb has snapped up a secretive AI startup named GamePlanner.AI a deal worth $200 million to reportedly develop the ultimate "travel concierge".

Little is known about GamePlanner.AI, or what technology it has been building to attract Airbnb's attention. Founded in 2020 by Adam Cheyer and Siamak Hodjat, the 12-person team has operated in stealth mode.

Cheyer and Hodjat are best known for building voice-activated chatbots. Cheyer has a long history in the industry and cofounded Siri, where Hodjat was a senior engineer, before [12]Apple bought it in 2010 . The pair moved to Apple before leaving to start Viv Labs, another startup that developed software that would later become the Bixby chatbot after being acquired by Samsung.

Airbnb's CEO Brian Chesky previously said he wants to use generative AI to build a travel concierge capable of learning about users' preferences over time so it can find the best homes for them to stay in, [13]according to CNBC. The home-rental marketplace biz gave little away in its [14]statement announcing the deal, only stating that GamePlanner.AI "will focus on accelerating select AI projects and integrating their tools."

Cheyer will join Airbnb as VP of AI Experience, and Hodjat will be a director. "A big part of what attracted us to Airbnb was our shared commitment to using AI to enable human connection," Cheyer said. "Like Brian, I believe that without great design and community-based intelligence, AI can only achieve a fraction of its potential."

Bard for teens

Google is opening up its Bard internet chatbot for teenagers to use.

In most countries, those 13 and older that have a Google account will be able to talk to Bard. Teenagers in Europe, Switzerland, and the UK, however, must be 18 or over. In South America, they must be at least 14 and over, and 15 in Korea. The web app, however, will only be available in English for those under 18.

Google wants teens to be able to interact with Bard to learn things like how to brainstorm ideas or ask questions about specific topics. They can also get Bard to help them learn skills or complete homework. Given an image of a math equation, Bard can, for example, explain the problem and give hints on how to solve it, Google [15]claimed .

[16]How 'AI watermarking' system pushed by Microsoft and Adobe will and won't work

[17]Microsoft rolls OpenAI's text-to-pic DALL-E 3 into Bing

[18]Google thinks $20M ought to be enough to figure out how or if AI can be used responsibly

[19]Bing AI feels like ChatGPT stuffed into a suit – not the future

There are more safeguards in place when teenagers use the software. Bard should avoid talking about inappropriate things that might be potentially harmful like illegal drugs. Its responses will automatically include a "G" icon younger users can click on to search Google, a feature the company calls "double-check response."

"We recognize that many people, including teens, are not always aware of hallucinations in LLMs. So the first time a teen asks a fact-based question, we'll automatically run our double-check response feature, which helps evaluate whether there's content across the web to substantiate Bard's response," Google [20]said in a blog post. ®

Get our [21]Tech Resources



[1] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/aiml&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=2&c=2ZVuQvNp4kxZLn6etJVZnWQAAAEk&t=ct%3Dns%26unitnum%3D2%26raptor%3Dcondor%26pos%3Dtop%26test%3D0

[2] https://twitter.com/Kyatic/status/1725120435644239889?ref_src=twsrc%5Etfw

[3] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/aiml&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=4&c=44ZVuQvNp4kxZLn6etJVZnWQAAAEk&t=ct%3Dns%26unitnum%3D4%26raptor%3Dfalcon%26pos%3Dmid%26test%3D0

[4] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/aiml&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=3&c=33ZVuQvNp4kxZLn6etJVZnWQAAAEk&t=ct%3Dns%26unitnum%3D3%26raptor%3Deagle%26pos%3Dmid%26test%3D0

[5] https://www.theregister.com/2023/10/31/judge_copyright_stabilityai_deviantart_midjourney/

[6] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/aiml&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=4&c=44ZVuQvNp4kxZLn6etJVZnWQAAAEk&t=ct%3Dns%26unitnum%3D4%26raptor%3Dfalcon%26pos%3Dmid%26test%3D0

[7] https://www.geekwire.com/2023/amazon-will-cut-several-hundred-more-jobs-on-alexa-team-as-it-discontinues-some-initiatives/

[8] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/aiml&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=3&c=33ZVuQvNp4kxZLn6etJVZnWQAAAEk&t=ct%3Dns%26unitnum%3D3%26raptor%3Deagle%26pos%3Dmid%26test%3D0

[9] https://www.theregister.com/2023/11/16/stability_ai_vp_quits/

[10] https://www.theregister.com/2023/11/15/house_of_lords_ai_copyright/

[11] https://www.theregister.com/2023/11/06/openai_gpt4_turbo_copyright_shield/

[12] https://www.theregister.com/2010/04/29/apple_siri/

[13] https://www.cnbc.com/2023/11/14/airbnb-acquires-ai-startup-for-just-under-200-million.html

[14] https://news.airbnb.com/airbnb-has-acquired-gameplanner-ai/

[15] https://blog.google/products/bard/google-bard-expansion-teens/

[16] https://www.theregister.com/2023/10/15/microsoft_adobe_ai_watermark/

[17] https://www.theregister.com/2023/10/04/microsoft_openai_bing_dalle3/

[18] https://www.theregister.com/2023/09/11/google_think_tanks/

[19] https://www.theregister.com/2023/03/15/bing_ai_first_look/

[20] https://blog.google/products/bard/google-bard-expansion-teens/

[21] https://whitepapers.theregister.com/



NoneSuch

And this is why corporations should not control AI output.

They will only let you see what they want you to see.

Hmm.

Tron

quote: And this is why corporations should not control AI output.

So you think governments will be more transparent and honest?

What planet are you from? I'd like to visit it one day.

Lee D

So, tell me:

If you have to control exactly what data you can feed it, double-check its responses with an automated system, and put in safeguards so it doesn't wander off topic, hallucinate, or talk about inappropriate things...

At which point is all of this just fancy heuristics ("human-written rules") and nothing more? Same way that Tesla still has to put in a human-managed blocklist of locations of where the AI goes totally wrong and they just say "Look, override and just drive straight here" (which appears to be a danger in itself!).

And how did it recreate anything Disney unless it was trained on Disney materials? It's not like someone sat and described in intricate detail what Mickey Mouse looked like and it created its own interpretation (which would be valid clean-room reverse engineering for most things anyway). It knew. It had seen it. And that's copyright infringement unless the imagery was licenced to be part of their training database. So who authorised that and what else is in there?

(I mean, I know the answers to all the above, it's pretty much rhetorical).

Catkin

"And that's copyright infringement unless the imagery was licenced to be part of their training database."

I believe this question is still bouncing around the courts and lobbyists, based on the process being one of pattern extraction; not to say it's definitely infringement-free, rather that it's not a question that's been answered definitively. There might be a secondary question of the source if Disney suspected the content were also pirated but it would be pretty hard to make this claim with movie posters, which are generally released for free distribution.

DJO

Lots of things are publicly available and publicly visible but you are not allowed to copy, bank notes are the most obvious example but artwork such as the Mona Lisa or some crappy film poster equally qualify.

In a lot of cases you can get away with "interpretations" and "derivative" works as long as you don't take the piss and try to pass off your work as original.

Doctor Syntax

"A big part of what attracted us to Airbnb was our shared commitment to using AI to enable human connection,"

And how big a part was that compared to the money?

My new AI killer app, exclusively revealed on El Reg.

Tron

It's a hardware/software combo delivering to the public the AI tech that they want and need the most.

It's a keyboard with an extra key - The F-Off key - that turns all the AI off.

If anyone out there has $200m to invest, I'm open to offers, bribes and honey traps.

You will not be elected to public office this year.