1.6.2023
Welcome to the trends in AI in May! This month we explore the latest insights and trends from the world of artificial intelligence. We'll cover controversies surrounding the impact of AI, the latest developments in European AI regulations, and new advancements from tech giants in the world of generative AI.
Cedric May
Chief Technology Officer
Geoffrey Hinton, the "godfather of AI," has resigned from his position as Google VP and engineering fellow. He expressed concerns over Google's lack of restraint with public AI releases, which he believes can lead to ethical problems. Hinton is worried about the potential for misinformation, job displacement, fully autonomous weapons, and AI models learning odd behavior from training data. Google's chief scientist, Jeff Dean, maintains that the company takes a responsible approach and is vigilant towards emerging risks.
''When something can do all kinds of things, I get a little bit worried. Because I know we won't be able to un-invent it and, you know, we did invent, for very, very good reason, the atom bomb in World War II'’
Renowned investor Warren Buffett, at the ripe age of 92, has expressed his concern regarding the potential threat posed by the development of AI. Despite the potential for AI to revolutionize various industries, Buffett remains skeptical about its ability to influence human behavior and thinking patterns. Charlie Munger, Vice Chairman of Berkshire Hathaway, shares Buffett's concerns and favors traditional intelligence instead. Geoffrey Hinton, a pioneer in the field of AI, similarly expressed his apprehension regarding the potential dangers of AI, citing it as a greater threat to humanity than climate change.
The European Union is edging closer to a historic ban on two controversial AI technologies - predictive policing and facial recognition - in public spaces. Two committees of MEPs have endorsed new AI rules that prohibit these technologies and introduce restrictions on generative models and emotion recognition. This regulation will be the world's first comprehensive AI law once approved. Tech companies like Microsoft and IBM, represented by The Software Alliance (BSA), have called for clarification, while civil liberties campaigners have praised the move. Privacy experts anticipate increased demand for their services as organizations will need to rely more on their privacy teams to operationalize AI and ensure governance.
Google CEO Sundar Pichai has agreed to work with EU lawmakers on a voluntary "AI Pact" while formal regulations for applying AI are still being worked on. The EU is launching an AI Pact "involving all major European and non-European AI actors on a voluntary basis" ahead of incoming EU legislation. The AI Pact may not have legal power, but it is a significant step towards international cooperation on rule-making. The EU recognizes the need for dedicated risk-based rules for AI and is doubling down on the 'Brussels effect' its digital lawmaking can attract by announcing a stop-gap AI pact now.
OpenAI's CEO Sam Altman recently toured Europe to discuss AI regulation and expand the company's reach. Altman expressed excitement about AI's versatility and potential to transform education. However, he also expressed concerns about the upcoming AI regulations set by the EU, which could cause OpenAI to leave Europe. Altman suggested that the EU could change the definition of general-purpose AI systems to account for AI tools with more than one application. Despite these challenges, Altman emphasized the benefits of AI and the need for a global regulatory framework. OpenAI plans to continue improving its models while remaining open to competition. Overall, Altman is optimistic about the future of AI and its positive impact on society.
Microsoft Designer, an AI-powered design tool similar to Canva, has launched in public preview. It generates designs for presentations, posters, digital postcards, invitations, and graphics for social media. Designer uses user-created content and OpenAI's DALL-E 2 text-to-image AI to ideate designs. The latest features include generating captions and hashtags for social media posts and creating animated visuals powered by AI. Designer is free during the preview period and can be accessed via the website and in Microsoft's Edge browser through the sidebar. It'll be included in Microsoft 365 Personal and Family subscriptions once generally available. Future features will include the ability to place an object in a specific spot and automatically fill in the rest of a picture.
Google Workspace was expertly designed for real-time collaboration, and now it's even better with AI-enhancements. Generative AI is seamlessly integrated across all Workspace apps to significantly boost productivity, including writing, organization, visualization, workflows, and meetings. This groundbreaking feature is called Duet AI for Google Workspace, which builds on the vision for Workspace and the launch of generative AI features in Gmail and Google Docs.
In March, OpenAI began developing plugins for ChatGPT to expand its capabilities. Over 70 third-party plugins are now being rolled out to ChatGPT Plus users, allowing access to the internet and various features. The plugins can be found in a user's settings and cover industries such as shopping, real estate, stocks, travel, and food. Users can install as many plugins as they want but are only allowed to use three at a time. The service costs $20 a month and gives access to OpenAI's ChatGPT-4 model and over 70 third-party plugins, which can parse through website content, visualize and analyze data, and even help learn foreign languages.
Meta has announced a new AI model called ImageBind that combines six types of data - visual, audio, text, depth, temperature, and movement - into a single multidimensional index. This research project is a significant milestone for the field of AI because it's the first to combine these six types of data into a single embedding space. The core concept is to link multiple types of data into a single index, paving the way for generative AI systems that can create immersive, multisensory experiences. With this technology, AI systems can cross-reference data, generating audio to match the video output and more. Meta has open-sourced the model, making this research accessible to all.
This week, Bing continued to release new features and respond to feedback. Some features will appear temporarily in your Bing chat experience as they experiment with them. These Release Notes will update you on which features have shipped as of today and are now rolling out to everyone.
New updates this week:
On May 24, AI-related stocks surged as Nvidia Corp forecast strong revenue growth and increased production of its AI chips to meet demand. Nvidia's stock rose up to 28% to trade at $391.50, increasing its market value by about $200 billion to over $960 billion, and extending its lead as the world's most valuable chipmaker. Other corporations related to AI, including Advanced Micro Devices Inc, Microsoft Corp, Google parent Alphabet Inc, C3.ai, and Palantir Technologies, also saw significant rises in their share prices. Nvidia's stock had already surged 109% in 2023 before this report.
Adobe Photoshop has recently launched Generative Fill, an innovative AI tool that allows users to quickly expand images and add or remove objects via text prompts. The tool is seamlessly integrated into individual layers within a Photoshop image file and offers three options for generating new objects or outpainting. Users can even leave the prompt blank for outpainting, but the system works even better when given some direction. Generative Fill is now available in beta and will be fully released in Photoshop later this year, showcasing Adobe's commitment to staying at the forefront of technological advancements. Adobe proudly claims that the model is only trained on content that the company has the legal right to use, ensuring it is safe for commercial use. Additionally, Generative Fill supports Content Credentials, a cutting-edge "nutrition label" system that attaches attribution data to images before sharing them online, letting viewers know if the content was created or edited using AI. Adobe has been successfully integrating AI-powered tools into its products for some time now, with Firefly already being employed in Adobe Illustrator to recolor vector-based images. Generative Fill can be downloaded as a desktop beta app or as a module within the Firefly beta app, allowing users to experience Adobe's latest breakthroughs with ease.
Geoffrey Hinton, the "godfather of AI," has resigned from his position as Google VP and engineering fellow. He expressed concerns over Google's lack of restraint with public AI releases, which he believes can lead to ethical problems. Hinton is worried about the potential for misinformation, job displacement, fully autonomous weapons, and AI models learning odd behavior from training data. Google's chief scientist, Jeff Dean, maintains that the company takes a responsible approach and is vigilant towards emerging risks.
''When something can do all kinds of things, I get a little bit worried. Because I know we won't be able to un-invent it and, you know, we did invent, for very, very good reason, the atom bomb in World War II'’
Renowned investor Warren Buffett, at the ripe age of 92, has expressed his concern regarding the potential threat posed by the development of AI. Despite the potential for AI to revolutionize various industries, Buffett remains skeptical about its ability to influence human behavior and thinking patterns. Charlie Munger, Vice Chairman of Berkshire Hathaway, shares Buffett's concerns and favors traditional intelligence instead. Geoffrey Hinton, a pioneer in the field of AI, similarly expressed his apprehension regarding the potential dangers of AI, citing it as a greater threat to humanity than climate change.
The European Union is edging closer to a historic ban on two controversial AI technologies - predictive policing and facial recognition - in public spaces. Two committees of MEPs have endorsed new AI rules that prohibit these technologies and introduce restrictions on generative models and emotion recognition. This regulation will be the world's first comprehensive AI law once approved. Tech companies like Microsoft and IBM, represented by The Software Alliance (BSA), have called for clarification, while civil liberties campaigners have praised the move. Privacy experts anticipate increased demand for their services as organizations will need to rely more on their privacy teams to operationalize AI and ensure governance.
Google CEO Sundar Pichai has agreed to work with EU lawmakers on a voluntary "AI Pact" while formal regulations for applying AI are still being worked on. The EU is launching an AI Pact "involving all major European and non-European AI actors on a voluntary basis" ahead of incoming EU legislation. The AI Pact may not have legal power, but it is a significant step towards international cooperation on rule-making. The EU recognizes the need for dedicated risk-based rules for AI and is doubling down on the 'Brussels effect' its digital lawmaking can attract by announcing a stop-gap AI pact now.
OpenAI's CEO Sam Altman recently toured Europe to discuss AI regulation and expand the company's reach. Altman expressed excitement about AI's versatility and potential to transform education. However, he also expressed concerns about the upcoming AI regulations set by the EU, which could cause OpenAI to leave Europe. Altman suggested that the EU could change the definition of general-purpose AI systems to account for AI tools with more than one application. Despite these challenges, Altman emphasized the benefits of AI and the need for a global regulatory framework. OpenAI plans to continue improving its models while remaining open to competition. Overall, Altman is optimistic about the future of AI and its positive impact on society.
Microsoft Designer, an AI-powered design tool similar to Canva, has launched in public preview. It generates designs for presentations, posters, digital postcards, invitations, and graphics for social media. Designer uses user-created content and OpenAI's DALL-E 2 text-to-image AI to ideate designs. The latest features include generating captions and hashtags for social media posts and creating animated visuals powered by AI. Designer is free during the preview period and can be accessed via the website and in Microsoft's Edge browser through the sidebar. It'll be included in Microsoft 365 Personal and Family subscriptions once generally available. Future features will include the ability to place an object in a specific spot and automatically fill in the rest of a picture.
Google Workspace was expertly designed for real-time collaboration, and now it's even better with AI-enhancements. Generative AI is seamlessly integrated across all Workspace apps to significantly boost productivity, including writing, organization, visualization, workflows, and meetings. This groundbreaking feature is called Duet AI for Google Workspace, which builds on the vision for Workspace and the launch of generative AI features in Gmail and Google Docs.
In March, OpenAI began developing plugins for ChatGPT to expand its capabilities. Over 70 third-party plugins are now being rolled out to ChatGPT Plus users, allowing access to the internet and various features. The plugins can be found in a user's settings and cover industries such as shopping, real estate, stocks, travel, and food. Users can install as many plugins as they want but are only allowed to use three at a time. The service costs $20 a month and gives access to OpenAI's ChatGPT-4 model and over 70 third-party plugins, which can parse through website content, visualize and analyze data, and even help learn foreign languages.
Meta has announced a new AI model called ImageBind that combines six types of data - visual, audio, text, depth, temperature, and movement - into a single multidimensional index. This research project is a significant milestone for the field of AI because it's the first to combine these six types of data into a single embedding space. The core concept is to link multiple types of data into a single index, paving the way for generative AI systems that can create immersive, multisensory experiences. With this technology, AI systems can cross-reference data, generating audio to match the video output and more. Meta has open-sourced the model, making this research accessible to all.
This week, Bing continued to release new features and respond to feedback. Some features will appear temporarily in your Bing chat experience as they experiment with them. These Release Notes will update you on which features have shipped as of today and are now rolling out to everyone.
New updates this week:
On May 24, AI-related stocks surged as Nvidia Corp forecast strong revenue growth and increased production of its AI chips to meet demand. Nvidia's stock rose up to 28% to trade at $391.50, increasing its market value by about $200 billion to over $960 billion, and extending its lead as the world's most valuable chipmaker. Other corporations related to AI, including Advanced Micro Devices Inc, Microsoft Corp, Google parent Alphabet Inc, C3.ai, and Palantir Technologies, also saw significant rises in their share prices. Nvidia's stock had already surged 109% in 2023 before this report.
Adobe Photoshop has recently launched Generative Fill, an innovative AI tool that allows users to quickly expand images and add or remove objects via text prompts. The tool is seamlessly integrated into individual layers within a Photoshop image file and offers three options for generating new objects or outpainting. Users can even leave the prompt blank for outpainting, but the system works even better when given some direction. Generative Fill is now available in beta and will be fully released in Photoshop later this year, showcasing Adobe's commitment to staying at the forefront of technological advancements. Adobe proudly claims that the model is only trained on content that the company has the legal right to use, ensuring it is safe for commercial use. Additionally, Generative Fill supports Content Credentials, a cutting-edge "nutrition label" system that attaches attribution data to images before sharing them online, letting viewers know if the content was created or edited using AI. Adobe has been successfully integrating AI-powered tools into its products for some time now, with Firefly already being employed in Adobe Illustrator to recolor vector-based images. Generative Fill can be downloaded as a desktop beta app or as a module within the Firefly beta app, allowing users to experience Adobe's latest breakthroughs with ease.
Read next
Convince yourself and redefine your success with little effort and resources for maximum results.
Your partner in redefining e-commerce, leveraging AI for hyper-personalized online experiences and superior product discovery.
