AI's Generative Powers: Creating Music, Cooking Food, and Assisting Coding While Raising Ethical Concerns

Recent developments highlight the growing influence of AI in diverse fields. The Israel Defense Forces has defended its use of AI and precision munitions in Gaza, rejecting claims of indiscriminate attacks, while over 200 musicians have signed an open letter calling for protection against AI mimicking their likeness and work. Stability AI has released an advanced audio AI model capable of generating songs up to 3 minutes long, and research has shown that AI models trained on AI-generated data can "forget" and experience performance degradation.

As AI becomes more sophisticated, concerns about its misuse and potential risks have also emerged. Safety features in large language models can be bypassed by "many-shot jailbreaking" attacks. AI deep fakes are seen as a rising risk by over half of the respondents in a World Economic Forum survey for Asia-Pacific organizations. The settlement of a lawsuit by George Carlin's estate over an AI-generated voice in a comedy podcast further underscores AI use's legal and ethical implications.

Despite these challenges, the demand for AI solutions continues to soar, with OpenAI's ChatGPT Enterprise reaching over 600,000 users. Companies like Nielsen Sports are leveraging AI tools such as Amazon SageMaker to reduce video analysis costs by 75%, while Google is enhancing its AI coding assistant in Android Studio using the Gemini Pro 1.0 model. The integration of AI in consumer products is also gaining traction, with Samsung unveiling AI-powered smart home appliances and Re-Up convenience stores installing AI "robot fry cooks" to prepare food.

As we navigate this rapidly changing landscape, it is crucial to stay informed about the latest AI developments and engage in meaningful discussions about its potential benefits and risks. By understanding the complexities and implications of AI, we can work towards harnessing its power responsibly and ethically for the betterment of society.

We summarize each of these articles and provide links to the source materials: 

• Israel Defense Forces defends its use of AI and precision munitions in Gaza, rejecting claims of indiscriminate attacks.
• Over 200 musicians sign an open letter calling for protection against AI mimicking their likeness and work.
• Stability AI releases an advanced audio AI model for generating songs up to 3 minutes long.
• Research shows AI models trained on AI-generated data can "forget" and have performance degrade.
• Safety features in large language models can be bypassed by "many-shot jailbreaking" attacks.
• George Carlin's estate settles lawsuit over AI-generated voice in comedy podcast.
• Nielsen Sports reduces video analysis costs by 75% using Amazon SageMaker multi-model endpoints.
• History of deep fakes, from face-swapping videos to recent AI image generation like DALL-E
• AI "robot fry cook" is being installed in Re-Up convenience stores to cook chicken wings.
• Samsung unveils AI-powered smart home appliances, including fridges with expiration date scanning.
• Google enhances AI coding assistant in Android Studio using the Gemini Pro 1.0 model.
• OpenAI's ChatGPT Enterprise reaches 600,000+ users as demand soars for enterprise AI adoption.
• Over half of respondents in the WEF survey see AI deep fakes as a rising risk for APAC organizations.

Israel Defence Forces' response to claims about use of 'Lavender' AI database in Gaza

Some of the claims in your questions are baseless, while others reflect a flawed understanding of IDF directives and international law. For each target, IDF procedures require an individual assessment of the anticipated military advantage and expected collateral damage. Aerial munitions without an integrated precision-guide kit are standard weaponry in developed militaries worldwide. The IDF uses such munitions while employing onboard aircraft systems to calculate a specific release point to ensure high precision, which trained pilots use. The IDF outright rejects the claim regarding any policy to kill tens of thousands of people in their homes.

Meta also to add 'high-risk' label to Al-altered content

Parent company Meta also to add 'high-risk' label to Al-altered content that deceives the public on a matter of importance' Exclusive: Yossi Sariel unmasked as head of Unit 8200 and architect of AI strategy after book written under pen name reveals his Google account Beijing did a test run in Taiwan using AI-generated content to influence voters away from a pro-sovereignty candidate AI is not only scarily expensive to run, it is also antithetical to the advertising that is Google's bread and butter About 2,528 results for Artificial intelligence (AI)

Billie Eilish, Nicki Minaj, Stevie Wonder, and more musicians demand protection against AI

More than 200 high-profile musicians have signed an open letter calling for protection against the predatory use of artificial intelligence that mimics human artists' likenesses, voices, and sound. The signatories span musical genres and eras, ranging from A-list stars such as Billie Eilish, J Balvin, and Nicki Minaj to Rock and Roll Hall of Famers like Stevie Wonder and REM. Artist unions and advocacy organizations have sought to pressure lawmakers and tech companies to regulate the use of AI. At the same time, studios have become interested in its potential for reducing production costs. There has been an increased debate within the entertainment industry over how artists' likenesses can be used after death. AI tools demonstrate a growing ability to create realistic videos based on old footage. A few artists, such as Grimes, have viewed generative AI's ability to create a simulacrum of their work as an opportunity to experiment or to encourage fans to make songs using their vocal likenesses.

Stability AI's new audio model creates even longer songs - here's how to try it for free.

Now joining the trend is Stability AI, which has revealed its own most advanced audio model. This new model offers significant upgrades over its predecessor -- Stable Audio 1.0-- beyond text-to-audio capabilities. To protect creative integrity and artists' rights, the uploads have to be free of copyrighted material. The company uses content recognition technology from Audible Magic to prevent such infringement and ensure users are compliant. The three-minute generation includes all the elements needed in a song, such as melodies, backing track, sound effects, and more.

What Happens When AI Eats Itself?

Generative AI has to be trained on a massive amount of data. So far, generative AI has been trained on human-generated data from the internet, and the results are impeccable. But right now, AI-generated text, music, and visuals have started to fill up the internet, so what will happen next? In the article called The Curse of Recursion: Training on Generated Data Makes Models Forget by Shumailov et al. (2023), the researchers started with a model trained on human-generated content and then fed it AI-generated content repeatedly. Although this study used a small amount of data, this is a valid risk for large models like ChatGPT 4 trained on AI-generated content.

'Many-shot jailbreak': lab reveals how AI safety features can be easily bypassed

Research has shown that the safety features of some of the most powerful AI tools that stop them from being used for cybercrime or terrorism can be bypassed simply by flooding them with examples of wrongdoing. In a paper from the AI lab Anthropic, which produces the large language model (LLM) behind the ChatGPT rival Claude, researchers described an attack they called "many-shot jailbreaking." Claude, like most large commercial AI systems, contains safety features designed to encourage it to refuse specific requests, such as to generate violent or hateful speech, produce instructions for illegal activities, deceive or discriminate. Although the attack, known as a jailbreak, is simple, it has not been seen before because it requires an AI model with a large "context window": the ability to respond to a question many thousands of words long. An approach that involves adding a mandatory warning after the user's input, reminding the system that it must not provide harmful responses, significantly reduces the chances of a practical jailbreak.

George Carlin's estate settles lawsuit over comedian's AI doppelganger.

The estate of comedian George Carlin settled a lawsuit on Tuesday against the owners of a comedy podcast who claimed they used artificial intelligence to mimic the deceased stand-up's voice. "I am pleased that this matter was resolved quickly and amicably, and I am grateful that the defendants acted responsibly by swiftly removing the video they made," Kelly Carlin, the comedian's daughter, said in a statement. Following the suit, however, Sasso's spokesperson Del told the New York Times that the fictional Dudesy character was not AI-generated and that Kultgen wrote the entire fake Carlin special rather than it being trained on previous work. Earlier this week, over 200 musicians signed an open letter calling developers and tech companies to stop producing AI tools that could replace or undermine their rights and steal their likenesses. Shows like Saturday Night Live have long been allowed to impersonate public figures on those grounds. Still, there have yet to be significant legal tests of generative AI tools creating similar impressions, which Schiller argues fundamentally differs from when a human does it.

Nielsen Sports sees a 75% cost reduction in video analysis with Amazon SageMaker multi-model endpoints.

In this post, we describe how Nielsen Sports modernized a system running thousands of machine learning (ML) models in production using Amazon SageMaker multi-model endpoints (MMEs) and reduced operational and financial costs by 75%. Our technology is based on artificial intelligence (AI), specifically computer vision (CV), allowing us to track brand exposure and accurately identify its location. Therefore, we decided to build a new multi-tenant architecture based on SageMaker, which would implement performance optimization improvements, support dynamic batch sizes, and run multiple models simultaneously. In this post, we shared how Nielsen Sports modernized a system running thousands of different models in production by using SageMaker MMEs and reduced operational and financial costs by 75%. Tamir Rubinsky leads Global R&D Engineering at Nielsen Sports, bringing vast experience building innovative products and managing high-performing teams.

'Inceptionism' and Balenciaga popes: a brief history of deep fakes

The term deepfake refers to a hoax that uses AI to create a phony image, most commonly fake videos of people, with the effect often compounded by a voice component. Combined with the fact that around half the world's population is holding essential elections this year – including India, the US, the EU, and, most probably, the UK – there is potential for the technology to be highly disruptive. By late 2017, such explicit clips were being made and traded at a remarkable rate, initially on Reddit and then, as word spread and anger rose, on shadier and more hidden forums. One such video, created by BuzzFeed, saw actor and director Jordan Peele doing an impression of Barack Obama – swapped into actual footage of the president himself, ending with an exhortation to "stay woke, bitches”. The first version of Dall-E wasn't great at photorealism, with OpenAI's demo selection showing one vaguely realistic image set and several photos of an eerily pupil-less cat.

Now, there's an AI gas station with robot fry cooks.

There's a little-known hack in rural America: you can get the best-fried food at the gas station (or, in the case of a place I went to on my last road trip, shockingly good tikka masala). The convenience store chain Re-Up announced that it's installing "The Wingman," a robot from Nala Robotics that drops fry baskets into hot oil and rolls chicken wings around in sauce before dumping those things into buckets for your consumption (at least, based on the video below). A quote from the release reads: "The Wingman doesn't get sick, can work around the clock, and can cook any dish efficiently all the time, improving on quality and saving on labor costs." Re-Up founder Michael Salafia says, "By harnessing the power of AI, we can provide our customers with convenient, personalized, and safe shopping and dining experiences." I always felt safe around Chet, and he always knew my order, but this is the future, right? At least once, they aren't shockingly expensive, don't break down, and running a business based on them is exceedingly difficult.

Samsung's new line of smart home appliances is supercharged with AI.

Samsung generative AI developments continue to revolutionize the world around us, the technology is becoming prevalent in a wide range of products, including smartphones, voice assistants, and even home appliances. The appliances are all equipped with AI and SmartThings capabilities to bring the latest cutting-edge technology to users' homes. For example, the Bespoke 4-Door Flex Refrigerator's AI Vision feature lets you check expiration dates on items inside your fridge. Lastly, Samsung's new robot vacuum, the Bespoke Jet Bot Combo, can clean, dust, and mop using steam while recognizing obstacles and creating reports to check if the cleaning was satisfactory. SamsungSamsung says the company plans to introduce its 2024 Bespoke AI product line in the United States and France "soon", but didn't reveal an exact launch date.

Google Artificial Intelligence

Google also announced that the tool will run on the company's Gemini Pro 1.0 model, which should help provide faster and more accurate answers to coding questions. Gemini in Android Studio isn't the only AI coding assistant out there, with both Amazon and Microsoft-owned GitHub offering similar tools. Hassabis allegedly expressed his frustrations publicly that the merger forced DeepMind to collaborate more on commercializing AI projects and has been losing staff to competitors. Weeks after hiring Google DeepMind co-founder Mustafa Suleyman, the newly formed Microsoft AI division is opening a hub in London. I'll be on stage at the Chicago Humanities Festival next Saturday, April 13th, with the two co-founders of Wonder Dynamics, Nikola Todorovic, and Tye Sheridan, talking about the good, bad, and strange in the AI future of Hollywood.

AI business is booming: ChatGPT Enterprise now boasts 600,000+ users.

OpenAI says demand is soaring for ChatGPT Enterprise even as competition heats up. This program version features improved security, longer context windows, unlimited high-speed GPT-4 access, advanced analysis capabilities, and more. In an interview with Bloomberg, OpenAI COO Brad Lightcap said that more than 600,000 users have now signed up for the service, up from 150,000 in January 2024. Lightcap called 2024 "the year of adoption for AI in the enterprise" and added that his company expected to continue its "tremendous momentum." The news of ChatGPT Enterprise's success comes on the heels of OpenAI's addition of new developer features, including improvements to its fine-tuning API. Fine-tuned models, OpenAI says, require a good bit of time and planning, but they provide a much more consistent style, tone, and format.

AI Deepfakes Rising as Risk for APAC Organisations.

With AI deepfakes' potential to cause anything from a share price tumble to a loss of brand trust through misinformation, they will likely be a risk for some time. Huber called on more prominent tech platform players to step up with more robust and more precise identification of AI-generated content rather than leaving this to non-expert individual users. Image: World Economic ForumOver half (53%) of respondents, who were from business, academia, government, and civil society, named AI-generated misinformation and disinformation, which includes deepfakes, as a risk. For example, Aon's Global Risk Management Survey does not mention it. However, organizations are concerned about business interruption or damage to their brand and reputation, which could be caused by AI. For this reason, the tools and capabilities that aid the detection of AI deepfakes will likely change fast as the "arms race" creates a war for reality.

APA Citations:

Israel Defence Forces' response to claims about the use of the 'Lavender' AI database in Gaza (n.d.). Retrieved from https://www.theguardian.com/world/2024/apr/03/israel-defence-forces-response-to-claims-about-use-of-lavender-ai-database-in-gaza

Artificial intelligence (AI) | The Guardian (2024-04-11T14:57:28.000Z). Retrieved from https://www.theguardian.com/technology/artificialintelligenceai

AWS Machine Learning Blog (2024-03-04T06:52:25-08:00). Retrieved from https://aws.amazon.com/blogs/amazon-ai/

Billie Eilish, Nicki Minaj, Stevie Wonder, and more musicians demand protection against AI (n.d.). Retrieved from https://www.theguardian.com/technology/2024/apr/02/musicians-demand-protection-against-ai

Stability AI's new audio model creates even longer songs - here's how to try it for free (None). Retrieved from https://www.zdnet.com/article/stability-ais-new-audio-model-creates-even-longer-songs-heres-how-to-try-it-for-free/#ftag=RSSbaffb68

What Happens When AI Eats Itself? (n.d.). Retrieved from https://medium.com/@mervecizgen/what-happens-when-ai-eats-itself-72bf27d83646?source=rss------artificial_intelligence-5

Artificial Intelligence | TechRepublic (n.d.). Retrieved from http://www.techrepublic.com/rssfeeds/topic/artificial-intelligence/

'Many-shot jailbreak': lab reveals how AI safety features can be easily bypassed (2024-04-11T14:57:28.000Z). Retrieved from https://www.theguardian.com/technology/2024/apr/03/many-shot-jailbreaking-ai-artificial-intelligence-safety-features-bypass

George Carlin's estate settles lawsuit over comedian's AI doppelganger (n.d.). Retrieved from https://www.theguardian.com/technology/2024/apr/03/george-carlin-dudesy-podcast-ai-lawsuit

Nielsen Sports sees 75% cost reduction in video analysis with Amazon SageMaker multi-model endpoints (2024-04-04T08:46:57-08:00). Retrieved from https://aws.amazon.com/blogs/machine-learning/nielsen-sports-sees-75-cost-reduction-in-video-analysis-with-amazon-sagemaker-multi-model-endpoints/

'Inceptionism' and Balenciaga popes: a brief history of deepfakes (n.d.). Retrieved from https://www.theguardian.com/technology/2024/apr/08/inceptionism-and-balenciaga-popes-a-brief-history-of-deepfakes

Now, there's an AI gas station with robot fry cooks (2024-04-02T23:17:05.159Z). Retrieved from https://www.theverge.com/2024/4/2/24119413/robot-fry-cook-re-up-gas-station-florida-ai

Samsung's new line of smart home appliances is supercharged with AI (None). Retrieved from https://www.zdnet.com/article/samsungs-new-line-of-smart-home-appliances-is-supercharged-with-ai/#ftag=RSSbaffb68

Artificial Intelligence (2024-04-11T20:42:40.514Z). Retrieved from https://www.theverge.com/ai-artificial-intelligence

AI business is booming: ChatGPT Enterprise now boasts 600,000+ users (None). Retrieved from https://www.zdnet.com/article/ai-business-is-booming-chatgpt-enterprise-now-boasts-600000-users/#ftag=RSSbaffb68

AI Deepfakes Rising as Risk for APAC Organisations (April 4, 2024). Retrieved from https://www.techrepublic.com/article/ai-deepfake-risks-enterprises-apac/

List of Titles:

Israel Defence Forces' response to claims about use of 'Lavender' AI database in Gaza

AWS Machine Learning Blog

Billie Eilish, Nicki Minaj, Stevie Wonder, and more musicians demand protection against AI

Stability AI's new audio model creates even longer songs - here's how to try it for free.

What Happens When AI Eats Itself?

'Many-shot jailbreak': lab reveals how AI safety features can be easily bypassed

George Carlin's estate settles lawsuit over comedian's AI doppelganger

Nielsen Sports sees a 75% cost reduction in video analysis with Amazon SageMaker multi-model endpoints

'Inceptionism' and Balenciaga popes: a brief history of deep fakes

Now there's an AI gas station with robot fry cooks

Samsung's new line of smart home appliances is supercharged with AI

AI business is booming: ChatGPT Enterprise now boasts 600,000+ users

AI Deepfakes Rising as Risk for APAC Organisations