#28 Deeptech Insights - Breaking AI News: Unlocking Next-Level Innovations, Projects and Tools
DeepTech Innovations Unlocked: This Week Our Expert News Analysis for Smarter Business and Investment Decisionss
❤️ If you like this format, please like this post.
🔎 Latest news on the AI Space
Google kicked off its annual I/O Event and has been in the AI headlines since then. If you missed the conference, here are the latest mind-blowing innovations that Google will offer
PaLM 2: Powering 25+ Google products with a next-gen language model sporting multilingual and coding prowess.
Bard: Available in 180 nations, 40 languages, boasting image I/O, coding upgrades, and app integration.
Search Generative Experience: Upgraded search delivering more info and context to your queries.
Search Labs: Play with novel Google products like generative search, Code Tips, and Add to Sheets.
Workspace Labs: New AI-aided features for Docs and Gmail, and a text-to-image tool for Slides.
Project Tailwind: An AI-first notebook, powered by your notes and sources.
MusicLM: Generate music with text prompts; the AI maestro is here.
Vertex AI: Meet Codey (text-to-code), Imagen (text-to-image), and Chirp (speech-to-text); your AI tool troika.
Gemini: Multimodal model with API integrations, still in training.
Immersive View in Maps: Multi-dimensional route visualization through AI and computer vision.
Magic Editor in Google Photos: AI-assisted precise image edits, as if by magic!
Generative AI Arrives to Photoshop. The popular AI image generator, Adobe Firefly is now available for everyone. Whether you're a beginner or a seasoned artist, let Firefly ignite your creativity.
OpenAI launches ChatGPT app for iOS
OpenAI has launched the ChatGPT app for iOS, enabling users to access ChatGPT on the go. The app syncs chat history, integrates speech recognition, and offers exclusive features for ChatGPT Plus subscribers. It allows instant answers, tailored advice, creative inspiration, professional input, and learning opportunities. The rollout will expand globally, with an upcoming release for Android users.Google unveiled a new audio model named SoundStorm. The model is capable of generating audio of the same caliber as AudioLM, but offers enhanced consistency. It operates 100x faster, and produces 30 seconds of audio in just half a second when using a TPU-v4.
Phoenix by Sanctuary AI: A humanoid general-purpose robot powered by Sanctuary AI’s Carbon AI control system designed to give Phoenix human-like intelligence and physical capabilities.
Exiii Inc. has developed a backpack with six robotic arms, Jizai Arms, that can be controlled, offering potential benefits for tasks and assisting disable individuals.
ImageBind by Meta AI: Multimodal AI model that can learn from six modalities, text, image/video, audio, depth, thermal and inertial measurement units to gain a holistic understanding of data.
Drag-GAN: user-friendly image-manipulation
It is an innovative tool designed for flexible and precise control over the manipulation of generated images. It allows users to "drag" any points of an image to precisely reach target points in a user-interactive manner. This approach is designed to provide users with precise control over where pixels move within an image, thus enabling manipulation of the pose, shape, expression, and layout of diverse categories such as animals, cars, humans, and landscapes. DragGAN is designed to produce realistic outputs even for challenging scenarios such as hallucinating occluded content and deforming shapes that consistently follow the object's rigidity. Both qualitative and quantitative comparisons demonstrate the advantage of DragGAN over prior approaches in the tasks of image manipulation and point tracking.
De-Aging Harrison Ford via SD
It’s an AI tool that leverages the power of generative adversarial networks (GANs) to perform age manipulation on images. It focuses on accurately synthesizing the age-progressed and age-regressed faces while preserving the identity of the individual. The uniqueness of this tool lies in its ability to perform the de-aging process by learning from a single input image instead of requiring a database of images at different ages. Its effectiveness has been demonstrated by de-aging the image of actor Harrison Ford, where it was able to perform the task in less than 6 minutes. The tool is currently undergoing further evaluation, with comparisons being made to other state-of-the-art approaches in the field
StableStudio: Stability AI announced StableStudio, an open source version of its DreamStudio application
Animation SDK: Stability AI released Stable Animation SDK, a text-to-animation open source toolkit. artists have the ability to use all the Stable Diffusion models, including Stable Diffusion 2.0 and Stable Diffusion XL, to generate animations
Adobe is releasing a powerful text-to-3d Image Composer. There is also a control widget that lets you edit/customize giving you more control.
AI can be used to understand animal emotions now. AI researchers reveal a facial scanning method able to detect if a cat is in pain. Approach: AI will track 'facial changes" linked to pain with 72% accuracy.
Artificial Nose by Microsoft: AI is slowing becoming more human. The Artificial Nose experiment is a smart device trained to recognize a variety of smells. It can identify the smell of bread, coffee, and more.
AI Doctor will soon become reality. Google’s PaLM 2 just scored 86.5% on the medical exam MedQA, shocking everyone.
Using AI to Detect Alzheimer’s: The paper explores using speech data and domain knowledge to detect Alzheimer's dementia, achieving 69% accuracy.
Humane is creating an AI-powered wearable device that lets users access computing power without being tethered to a smartphone or other device.
Meta and Mark Zuckerberg announce 'Massively Multilingual Speech.' This open-sourced project can identify 4000 languages online through software.
🔍 AI Radar
OpenAI’s CEO Sam Altman delivered a quite interesting testimony to the US Senate.
Together announced a $20 million seed round to accelerate their work in open source and cloud AI platforms.
Hippocratic raised $50 million to accelerate their vision of creating LLM models for healthcare.
Union AI raised $19.1 million to accelerate its AI-first ETL platform.
Yellow AI announced YellowG, a new conversational platform for workflow automation.
Procurement platform Zip disclosed a $100 million fundraise to incorporate AI capabilities.
AMP Robotics has recently launched an AI-powered garbage-sorting robot called the AMP Cortex system. This will revolutionize the recycling industry (for good). They just raised an $8M Series C funding round
💻 CooI AI Tools / Startup
Shape-e from OpenAI: Generate 3D objects conditioned on text or images
Personal AI - Pi: A chatbot which is designed to be a kind and supportive companion offering conversations, friendly advice, and concise information in a natural, flowing style.
Hugging Face launches open-source version of ChatGPT: Hugging Face, a well-known player in the open-source AI development arena, has unveiled open-source 30B chatbot alternative to ChatGPT, named HuggingChat. The new platform boasts a user interface that lets users interact with an open-source chat assistant called Open Assistant
Cursor: Cursor is an AI-powered programming editor that generates and edits code, and offers other useful features like chat (ChatGPT-style interface that understands your current file). Open-source alternative to GitHub Copilot.
Segment Anything Model (SAM): a new AI model from Meta AI that can "cut out" any object, in any image, with a single click · SAM uses a variety of input prompts.
Recast: It's a Chrome extension where you simply instruct it to recast any article, and it'll condense it down into a shorter piece and then read it out like a podcast. There are multiple hosts and it sounds like one person interviewing another about the article. It really makes listening to articles much more engaging.
Blend Studio: Create professional product photos and designs in two clicks
Personified- Create a ChatGPT with your organisation's data
PrivateGPT: Interact privately with your documents using the power of GPT, 100% privately, no data leaks
OpenChatKit: The Open Source Alternative to ChatGPT
LeMUR by Assembly AI: Leveraging LLMs to transcribe up to 10 hours worth of audio content (~150K tokens) with a single line of code.
Exemplary AI - Transform audio/video into content, summaries & insights.
Lalal.ai: This tool uses a neural network system called Phoenix to automate audio source separation, extracting elements such as vocals, music, or specific instrumental tracks from any audio or video content.
ArchitectGPT: it can help you create stunning visuals of your home or property.
KIIT: ChatGPT with live audio/video capabilities that can answer questions, take meeting notes, and act as a multilingual translator
Bizway: Turn your ideas into business plans in minutes, provides customizable roadmaps, auto-generating and completing task
Vondy AI: 100+ generative AI tools all bundled into one API designed for developers.
💡Spotlight: Controlling Your Computer With Your Thoughts
Neuralink, founded by Elon Musk, is pushing the boundaries of technology with its brain-computer interface (BCI). The company aims to create a groundbreaking implant that allows individuals to control computers and devices with their thoughts. While the potential benefits are immense, concerns about safety, ethics, and privacy surround this emerging technology.
The Power of Neuralink's Brain-Computer Interface: Neuralink's BCI holds promise for people with paralysis and neurological disorders, offering them the ability to interact with digital systems using their thoughts. By implanting electrode-filled threads, thinner than a human hair, directly into the brain, the BCI aims to restore independence and empower individuals who have lost motor function.
Safety and Regulatory Challenges: The implantation procedure, involving a minimally invasive surgery conducted by a robot, is said to carry minimal risk and take only an hour or two. However, the long-term effects and potential risks on humans are yet to be fully understood. Neuralink's initial application for human trials was rejected by the FDA due to safety concerns, highlighting the need for rigorous evaluation and compliance with regulatory standards.
Ethical Concerns and Investigations: Neuralink is under scrutiny for its treatment of animals during preliminary trials and allegations of mishandling pathogenic materials. Ethical considerations surrounding the company's practices must be addressed to ensure the responsible development and implementation of the technology.
Security and Privacy Considerations: As the BCI technology offers a direct connection between the brain and digital systems, concerns over security and privacy have arisen. Users worry about potential data breaches and unauthorized access to their thoughts and personal information. Establishing robust privacy regulations will be essential to alleviate these concerns and ensure the secure use of the technology.
The Future of Neuralink and Personal Choice: Neuralink's path to regulatory approval remains uncertain. The decision to undergo a transformative procedure like the BCI implant will ultimately depend on individual risk tolerance, medical necessity, and ethical considerations. While the technology opens up exciting possibilities for human-machine interaction, it is crucial to thoroughly evaluate the risks and benefits before making a personal choice.
Neuralink's brain-computer interface has the potential to revolutionize human-machine interaction, offering new possibilities for individuals with paralysis and neurological disorders. However, safety, ethics, and privacy concerns need to be carefully addressed before widespread implementation. As the world eagerly awaits further developments, the decision to embrace such a transformative technology will ultimately rest with individuals, considering the risks, benefits, and ethical implications involved.
Would you get the implant if it was brought to market?
✨ That’s all for today. Thanks for reading ! Stay tuned for our next article coming up end of week with our Deeptech Insights Newsletter.
Much love Deeptechers!👋💖