Open AI has created a voice model that can generate realistic voices from a single 15-second sample. The audio generated is of high quality, even surpassing the capabilities of tools like Eleven Labs.

While this technology presents exciting possibilities, Open AI acknowledges the potential risks. They are not making the model publicly available, as they want to ensure it is not misused.

To address the challenges, Open AI is encouraging steps such as:

• Phasing out voice-based authentication as a security measure
• Exploring policies to protect the use of individuals' voices
• Educating the public about the possibility of deceptive AI-generated content
• Accelerating the development and adoption of techniques for tracking the origin of audiovisual content

By taking a responsible approach, Open AI aims to navigate the complexities of synthetic voices and ensure this technology is used in a manner that benefits society while mitigating potential harms.

Open AI has added a new inpainting feature to Dolly 3, their image generation model. This allows users to select an area of a generated image and prompt the model to make changes to that specific region.

For example, if you generate an image with Dolly and are not satisfied with certain elements, like the wheels on a hoverboard, you can highlight those areas and prompt the model to "remove the wheels". Dolly will then regenerate that part of the image, attempting to remove the wheels as requested.

This inpainting capability gives users more control and flexibility when working with Dolly-generated images. It allows for iterative refinement and editing, rather than having to start from scratch. However, the feature may still require a few tries to get the desired result, as the model doesn't always perfectly execute the requested changes.

Overall, the addition of inpainting is a useful enhancement to Dolly 3, empowering users to fine-tune and customize the images produced by the model.

Stability AI this week announced Stable Audio 2.0, which includes a new feature called audio to audio generation. With this feature, you can hum or make instrument sounds with your mouth, and the AI will attempt to replicate that with generated audio.

Here's an example:

[Music]

While the feature is interesting, the generated music is not yet at the level of quality or excitement that we see from tools like Jukebox. The songs generated by Stable Audio 2.0 are still fairly basic and lack the complexity and richness of human-composed music.

Overall, the audio to audio generation in Stable Audio 2.0 is a step forward, but it still has room for improvement before it can truly compete with professional music production.

A group of musicians, including Nicki Minaj, Billy Eilish, and Katy Perry, have signed a letter against the irresponsible use of AI in the music industry. The key points from the letter are:

• The signers call on AI developers, tech companies, platforms, and digital music services to cease the use of AI that infringes upon and devalues the rights of human artists.

• However, the letter acknowledges that when used responsibly, AI has enormous potential to advance human creativity and enable new and exciting experiences for music fans.

• The letter expresses concern about the "predatory use of AI to steal professional artist voices and likenesses, violate creator rights, and destroy the music ecosystem."

• The signers state that the "assault of human creativity must be stopped" and they must protect against the misuse of AI that could harm artists and the music industry.

• Overall, the letter takes a cautious stance, recognizing both the potential benefits and risks of AI in music, and calling for responsible development and use of the technology to protect the rights and livelihoods of human artists.

Crea AI, the AI-powered image generation tool, has recently rolled out a new "Image Blend" feature. This feature allows users to upload multiple images, adjust the weight of each image, and see the resulting image blend in real-time.

For example, the Crea AI team demonstrated this feature by uploading three images with the prompt "fish made out of porcelain." As they adjusted the weight of the various images, the output image dynamically blended the elements from the different source images.

This new capability is quite impressive, as it enables users to seamlessly combine various visual elements into a single, cohesive image. The real-time blending allows for quick experimentation and refinement, making Crea AI an even more powerful tool for creative expression and ideation.

The ability to blend multiple images opens up new possibilities for Crea AI users, allowing them to create more complex and nuanced visual compositions. This feature further demonstrates the ongoing advancements in AI-powered image generation, pushing the boundaries of what is possible with this technology.

The rapid advancements in AI technology have raised concerns about the potential impact on jobs. Many big tech companies are now coming together to address this issue through the ICT Workforce Consortium.

The consortium, which includes companies like Cisco, Google, Microsoft, and IBM, aims to find ways to prevent people from losing their jobs due to the rise of AI. The goal is to figure out how to keep people employed as AI continues to evolve and potentially automate certain tasks.

This is an important initiative, as there have been reports of major tech companies like Google laying off thousands of employees, with some speculation that AI was a contributing factor. However, the reality is that these layoffs were more about companies restructuring to focus more on AI development, rather than AI directly replacing human workers.

Nevertheless, the potential for AI to disrupt traditional job roles is real, and the ICT Workforce Consortium is a proactive step towards addressing this challenge. By bringing together industry leaders, the consortium aims to develop strategies and solutions to help workers adapt to the changing job landscape and ensure that the benefits of AI are shared broadly, rather than concentrated among a few tech giants.

As AI continues to advance, it will be crucial for companies, policymakers, and workers to collaborate and find ways to harness the power of this technology while also protecting jobs and livelihoods. The ICT Workforce Consortium represents an important step in that direction.

A court in Washington has banned the use of AI-enhanced video evidence. The concern is that AI-based upscaling tools, like Topaz Labs, do not actually improve the quality of the video footage. Instead, the AI tries to guess the pixels in between, potentially adding or altering details that were not originally present in the footage. This raises issues with the reliability and authenticity of the evidence, making it unsuitable for use in court proceedings. The ruling highlights the need for caution when using AI-enhanced media, as the technology may inadvertently introduce inaccuracies or distortions, compromising the integrity of the evidence.

The Apple Vision Pro now allows users to interact with others in virtual environments. Some key features include:

• Virtual Presentations: Users can appear as a floating disembodied head with hands to present content and collaborate with others in virtual spaces.

• Virtual Games: Users can play virtual board games, chess, and other interactive experiences with remote participants. This allows for shared gaming experiences in the headset.

• Watch Movies Together: Users can watch movies or other content together, with each participant appearing as a virtual presence in the shared environment.

The ability to interact with others in the Apple Vision Pro represents an important step forward in making the headset a more social and collaborative platform, moving beyond the isolated experience of previous virtual reality devices. These new features enable users to engage with friends, colleagues, and others in shared virtual experiences.

A company out in India has created a scooter called the Ola Solo, which claims to be the first fully autonomous electric scooter. The idea is that you can jump on this electric scooter and it will take you where you need to go, without a driver. While this is an interesting concept, the author is not sure if they would feel comfortable riding an autonomous scooter, at least not the first version. It will be something to follow and see if it ever comes to fruition.

Additionally, in Phoenix, Wejo vehicles are now delivering Uber Eats. Wejo is a self-driving car company, and the idea is that you can place an Uber Eats order, and a self-driving car will come pick up the food from the restaurant and deliver it to your house. You can even opt-out during checkout if you prefer a human delivery driver. This is another fascinating application of self-driving technology.

The upcoming season 6 of the Netflix reality show "The Circle" will feature an AI catfish as one of the contestants. In this dating-style show, participants communicate with each other through a social media platform without ever meeting in person. Some contestants even create fake personas, known as "catfishes", to try and win the game.

For the first time, an AI bot will be one of these catfish characters, pretending to be a human named "Max". It will be fascinating to see how the other contestants interact with and respond to this AI-powered participant. Will they be able to detect that "Max" is not a real person? And if they do, will they still try to form a connection with the AI?

This experiment will provide interesting insights into how people perceive and engage with AI in a social setting. It will be exciting to see if the AI is able to successfully navigate the social dynamics of the show and potentially even form relationships with the human contestants. The outcome of this AI's performance could have broader implications for the future integration of AI into our social lives.

The rapid advancements in AI technology are truly astounding. From the overwhelming AI landscape to the ambitious data center project by Microsoft and OpenAI, the field is progressing at an unprecedented pace.

The development of OpenAI's voice model, which can generate realistic voices from a 15-second sample, highlights the impressive capabilities of modern AI systems. However, the responsible approach taken by OpenAI in not publicly releasing this technology due to potential misuse is commendable.

The introduction of new features like inpainting in Dolly 3 and the ability to use tools within Claude 3 showcase the continuous improvements in AI capabilities. Apple's efforts to integrate AI into its products, such as the Realm language model for Siri, also demonstrate the industry's commitment to advancing this technology.

While the potential benefits of AI are vast, the concerns raised by musicians about the responsible use of AI in the music industry are understandable. The formation of the ICT Workforce Consortium by tech giants to address the impact of AI on jobs is a positive step towards addressing these challenges.

The emergence of innovative AI-powered applications, such as the autonomous electric scooter and the AI-powered delivery service by Wejo, highlights the diverse applications of this technology. The inclusion of an AI chatbot as a cast member in the Netflix reality show "The Circle" also showcases the integration of AI into various forms of entertainment.

Overall, the AI landscape continues to evolve rapidly, with both exciting advancements and thought-provoking ethical considerations. As the technology progresses, it will be crucial to navigate these challenges and ensure the responsible development and deployment of AI for the benefit of society.