AI Updates

News:

  • OpenAI is committed to developing safe and broadly beneficial AI. Today we are sharing preliminary insights and results from a small-scale preview of a model called Voice Engine, which uses text input and a single 15-second audio sample to generate natural-sounding speech that closely resembles the original speaker. It is notable that a small model with a single 15-second sample can create emotive and realistic voices.

  • Is Sora stealing data: Despite a lack of concrete evidence that OpenAI is using YouTube data to train Sora, its text-to-video model, YouTube CEO—Neal Mohan—has declared that it will be a “clear violation” of its terms of service if it has. He states that when a creator uploads their work to YouTube, they expect their terms of service to be met, and having their content used by a third party is “a violation” of this. OpenAI CTO (Mira Murati) said Sora was trained on "publicly available data and licensed data" but didn't know if that included videos from YouTube. OpenAI has previously admitted using copyrighted data to train its models because it’s “impossible” to build the technology without it.

  • Cisco to the rescue. Some of the biggest global tech companies—including Google, IBM, and Microsoft—have formed a consortium to address the impact of AI-related job losses. The consortium (AI-Enabled Information and Communication Technology Workforce Consortium) will upskill or reskill employees who have lost (or could lose) their jobs. This comes as estimates show that 30% of back-office jobs will be automated by AI, causing 25% of employees to expect AI-related layoffs, with 4,000 employees already replaced.

  • OpenAI is open to help expand its custom model program. OpenAI is expanding its Custom Model program, which gives Enterprises access to OpenAI researchers who can help train AI models for specific use cases and applications. With dozens enrolling, OpenAI plans to introduce “assisted fine-tuning” and “custom-trained models” to further bolster model performance on particular tasks. This extended program saw law firm Harvey create a custom AI-powered tool incorporating billions of legal texts and attorney feedback to speed up case preparation.

  • New York defends chatbot giving illegal advice. New York City Mayor (Eric Adams) has defended ‘MyCity’—the first city-wide chatbot to provide business owners “trusted information”—even after it gave wrong and illegal advice. It wrongly advised employers that they could take their workers’ tips and claimed that they didn’t need to give their employees notice if their schedules changed. MyCity is still live, with Adams proclaiming that it was a pilot program—launched in October—so “you need to put it into the real environment to iron out the kinks.”

  • Pay Google to Search? Google is considering charging users for content generated by SGE (Search Generative Experience), a search experience that uses AI to provide users with overviews of search topics. Google’s AI-powered search features could become part of its existing subscription services, with its traditional search engine remaining free. It’s thought that although subscribers will pay for AI-generated search response summaries, they will still receive ads, as they will when using the free Google Search. Although Google already charges users for Gemini and extra storage, this move marks the first time it’s put one of its core search products behind a paywall.

  • Musk is raising the bar: Elon Musk is dramatically raising his AI engineer salaries because OpenAI has been “aggressively recruiting AI engineers with massive compensation offers.” He nearly lost key AI scientist Ethan Knight to OpenAI but convinced him to stay with a higher salary and a move to his xAI startup: “It was either xAI or them.” This comes as the battle for AI talent heats up, with Mark Zuckerberg sending personal recruitment emails and offering jobs without interviewing candidates.

  • Will the launch of Sora be “Diffused”: OpenAI’s Sora has a new (more accessible) rival. OpenAI’s Sora—set to be used in Hollywood—has a competitor (Higglesfield) that has created a similar AI-powered video generation tool (Diffuse) but for the masses. Like Sora, Diffuse is powered by a text-to-video model—with a prompt editor that users can use to describe scenes—which it will use to generate realistic film footage. But, whereas Sora is probably more for high-end creatives, Diffuse is for “creators of all types” and could be an alternative for everyday users or social media marketers.

  • Understanding AI safety: The UK and US AI Safety Institutes (established in November 2023) have signed an agreement (called The Memorandum of Understanding) to test and monitor advanced AI models for safety risks, marking the first bilateral agreement of its kind. The agreement states that both countries will collaborate on evaluating the safety of AI tools and systems, creating a common approach to testing AI safety. It also establishes that they will share employees, exchange information, and perform a joint testing exercise on a publically available AI model. This comes after companies (such as Google and OpenAI) agreed to voluntarily test AI systems, an agreement backed by 10 countries, including the US and UK.

  • Don’t mess with music: Over 200 music artists—including Stevie Wonder, Billie Eilish, Katy Perry, and Pearl Jam—have signed an open letter asking AI leaders to “protect against the predatory use of AI.” The letter (organized by the Artists' Rights Alliance) asks tech firms not to develop AI music-generation tools "that undermine or replace the human artistry of songwriters and artists." It states that the way artists' voices are being used to train AI models is "an assault on human creativity" and warned it “will destroy the music ecosystem".

  • Beyonce makes a stand against AI In the middle of a press release to launch her new album—“Cowboy Carter”—Beyonce made a statement against the increasing influence of AI in the music industry. She expressed that authenticity is essential when creating her music and emphasized her preference for real instruments over AI tools and digital manipulation. This proves her commitment to preserving the artistry of music production at a time when AI music generators can create tracks and emulate artists’ voices without consent.

  • Apple’s new AI model beats GPT4 Apple researchers have revealed that their new small language model (called ReALM)—designed to make voice assistant (Siri) smarter—has outperformed GPT-4. ReALM can “see” on-screen information and understand context, so if you were looking online and wanted Siri to call a company, Siri would “see” the number and make the call. ReALM outperformed GPT-4 regarding nuanced queries, proving it better understands context, can process onscreen content, and can respond accordingly.

  • Sam steps aside: Sam Altman, has transferred control of OpenAI’s VC fund—the OpenAI Startup Fund, which invests in early-stage AI businesses—to fund management partner, Ian Hathaway. The change was made to address potential conflicts of interest, as (while being OpenAI’s CEO) Altman established the fund, raised capital, and made all investment decisions. Hathaway—who has been a partner since 2021—was previously overseeing the fund's accelerator program, and led investments in companies such as Harvey and Ambience Healthcare.

Be intentional, have a plan:

  • DoraMaria spoke about using a good framework on Saturday https://app.fireflies.ai/view/DoraMaria-AI-Frameworks-mp3::Ot56AcQlBxR4y6Zf

  • CREATE: Context request, examples, and audience. 

  • RTF: Role, Task, Format

  • TAG: Task, Action Goal

  • BAB: Before, After, Bridge

  • CARE: Context Action Results Example

  • RISE: Role, Input, Steps Expectation

  • https://www.clubhouse.com/@dmabreu

  • LLMs from scratch (beginners guide to Large Language Models) https://www.youtube.com/watch?v=lnA9DMvHtfI&t=2s

    Take Massive Action:

  • The Hero’s Journey

  • Finish The Story

    Realtor Tools:

  • Focus on generating leads and becoming a true SME

  • agentpulse.ai

  • write.homes

  • tableau.com

  • domo.com

    Creator Tools:

  • Claude: A conversational AI platform designed for nuanced and contextually aware interactions, aiming to provide human-like conversation experiences. https://claude.ai/chats

  • ChatGPT: A platform offering access to OpenAI's GPT model, tailored for engaging in conversational responses, providing information, and generating text-based content. https://chatgpt.com/

  • Pika Labs: A creative platform focused on AI-driven art creation, allowing users to explore and create digital artworks with the assistance of artificial intelligence. https://pika.art/

  • Runway: A creative toolkit powered by AI, enabling users to apply machine learning models to video, image, and text projects for innovative content creation. https://runwayml.com/

  • Leonardo: An AI platform (not widely known as of my last update, so this description is speculative) likely aimed at enhancing digital art creation or providing AI-based tools for creative processes. https://leonardo.ai/

  • Storyblocks: A stock media platform offering royalty-free videos, images, and audio clips for content creators to use in their projects. https://www.storyblocks.com/

  • Dalle-3 on Bing: Microsoft’s integration of OpenAI's DALL-E model into Bing allows users to generate unique images based on textual descriptions through a web interface. https://www.bing.com/images/create

  • Ideogram: An AI-driven platform focused on generating ideograms or visual symbols that represent ideas or concepts, facilitating creative visual communication. https://ideogram.ai/t/explore

  • MusicFX: A Google experiment that allows users to explore the creation of music using AI, part of Google's AI Test Kitchen, focusing on experimental AI applications in music. https://aitestkitchen.withgoogle.com

  • Mubert: An AI-powered platform for generating unique music streams, allowing creators to produce music through algorithmic composition. https://mubert.com/

  • Suno: An AI tool focused on enhancing voice communication by offering real-time, AI-driven voice analysis and improvement features (based on the description, this is speculative). https://www.suno.ai/

  • ElevenLabs: Offers AI technology for creating realistic voice synthesis and cloning, enabling high-quality voice generation for various applications. https://elevenlabs.io/

  • Adobe Speech Enhancer: A tool designed to improve the quality of audio recordings, especially in podcasting, by removing noise and enhancing speech clarity. https://podcast.adobe.com/enhance

  • Timebolt: A video editing software that automates the process of cutting out silences from video content, making content creation more efficient. https://www.timebolt.io

  • Descript: A multi-tool platform for audio and video editing that features transcription, text-to-speech, and media editing capabilities, aimed at content creators. https://www.descript.com/

  • HeyGen: Provides AI-driven tools for generating written content, aiming to assist in the creative writing process with the help of artificial intelligence. https://heygen.com/

  • Opus Clip: A platform (as of my last update, not widely recognized, so description is speculative) likely focused on video editing or creation tools enhanced by AI. https://www.opus.pro/

  • TubeBuddy: A browser extension and mobile app designed to help YouTube creators optimize their videos, manage their channels, and grow their audience. https://futuretools.link/tubebuddy-com

  • Invideo: An online video creation platform offering tools and templates for creating professional-quality videos for marketing, social media, and more. https://invideo.io/

  • LTX Studio: Specializes in leveraging AI for text-to-video generation, allowing users to create video content from written narratives. https://ltx.studio/