Last week was packed with groundbreaking developments pushing the boundaries of artificial intelligence(AI) and robotics. The future is unfolding, from biocomputers made of human brain cells to advanced humanoid robots. We’ll get into the latest AI models revolutionizing image generation, explore the legal battles shaping the AI landscape, and unveil cutting-edge technologies and updates that seem like science fiction but are now a reality. In this article, we will cover all the significant updates and trends in AI, along with some of the latest AI tools and new jobs listed in the field of AI.
Table of Contents
Trending AI News of the Week
Created AI using human brain cell
FinalSpark, a Swiss startup, has launched a new service that allows scientists and researchers to rent biocomputers made from human brain cells. This startup uses clumps of human brain cells that can be used to compute for up to 100 days. To train these AI models, the startups use dopamine for positive reinforcement and electrical signals for negative reinforcement. FinalSpark also claimed that these biocomputers could be 100,000 times more efficient for AI training.
Also Read – Top 5 AI Video Generators: The Future of Marketing
Ideogram 2.0 new updates
Ideogram is an AI tool that generates images with expertise in generating text, which many models have failed to do. The tool released its “V2” with many new features, including new image styles like Realistic, Design, 3D, and Anime. The latest model improved human textures, such as hands, hair, facial expressions, etc., and significantly enhanced text rendering. Users can generate around 40 images in a day at no charge.
Also Read- 7 Best FREE AI Chatbots That Will Blow Your Mind
xAI releases the Grok 2 in early access
xAI released the early AI model Grok 2, a powerful AI model trained on the enormous data of the X and using the real-time data from X to generate output. The model also uses Flux.1 to generate unfiltered images. As X is the world’s largest microblogging app, it allows Grok 2 to get the information in real time. This model is currently available for some premium X users and will soon be available for others. Also, there are rumors that Grok 3 will be launched in December 2024.
Also Read- Claude AI – Better than GPT-4?
Disney developed a new AI System.
Disney Research Scientists and ETH Zurich developed a new AI for virtual characters. This new system can generate realistic movement of virtual characters and robots using physics principles from simple text or image inputs. This system uses a two-stage approach, which allows it to handle various motions, such as walking and complex acrobatics of the virtual characters and robots. It can also transfer the movements of humans to robots while maintaining the principles of physics. This technology would help the various robotics companies and startups to fasten their development.
Also read- Best GPT-4 Plugins: Use ChatGPT like a pro
Companies sued LLMs for copyright infringement.
A lawsuit has been filed against Anthropic and OpenAI for committing the crime of copyright infringement. These companies developed the world mainly using chatbots, Claude and ChatGPT, and these are large-scale AI models trained on massive datasets. The lawsuit claimed that these companies use the content of pirated copies of copyrighted books. However, these companies are protected under the fair use of doctrine U.S. laws, which allows them to limit the use of copyright materials. Let’s see what the court’s verdict will be in this matter.
Also Read: Nightmare AI Review: The AI Image Upscaler
Microsoft launched a new AI model.
Microsoft recently launched a new AI model, “Phi-3.5 MoE.” It is the latest Microsoft Phi series of compact AI models member. It is also a small and compact AI model with an efficient architecture. This model uses a new mixture of experts(MoE) approach, which makes it capable of beating many LLMs on the reasoning benchmarks. Phi-3.5 MoE has outperformed renowned models such as Llama 3 8B, Google’s 2 9B, and OpenAI’s GPT-4o mini. This hints that the future of AI models will be small and compact and run directly on mobile devices.
Also Read: Best AI Documentaries To Watch
AGIBOT launched a robot family
A Chinese-based robotics startup, AGIBOT, released a family of five advanced humanoid robots. This series has five models, which are both bipedal and wheeled, designed for a variety of tasks from households to industries. The flagship model, Yuanzheng A2, has a height of 175 cm and a weight of around 55 kg. Another robotics company, Unitree, showcased its new production-ready robot, G1, with new features.
Also Read: OpenAI launched a new LLM, “GPT-4o mini”
Luma launched Dream Machine 1.5
Luma Labs just released Dream Machine 1.5, a successor to its existing AI video generation model. This new AI model can generate high-quality text-to-video videos, more brilliant prompt understanding, and better image-to-video videos. Dream Machine 1.5 can generate high-quality videos of 5 seconds from text and image inputs. This new model comes with the expertise in creating smooth motion, cinematography, and dramatic shots. Currently, it is available to try for free.
Click here to try
Trending AI Tools
- GigaBrain (Click here to try)
It is an AI tool that scans the billions of discussions on Reddit and brings you critical posts. - Graphy (Click here to try)
It is an AI tool that transforms rough data into visual presentation. - Tome (Click here to try)
It is an AI tool that helps users in the sales area. - MYND (Click here to try)
It is an AI tool that helps users to overcome mental blocks. - Bayesline (Click here to try)
It is an AI tool that helps investment managers with financial analytics. - MolyPix (Click here to try)
It is an AI tool that generates easy-to-edit graphic designs from text.
Also Read: Best AI Humanoid Robots
Tool of the Week
Last week, I was making a presentation for a new product, and during the making, I needed some realistic images to make the presentation more visually appealing. So, I searched the internet and didn’t get the right image. Then I thought, why don’t we generate the images with the help of AI? After much research, I found a new AI model, Fal AI, which can generate ultra-realistic images simply by describing the vision.
Steps
- Visit the website- Click here.
- In the Prompt box, write a detailed prompt for the desired image.
- Click More for Additional Settings to fine-tune parameters like image size, number of steps, and guidance scale.
- Click on “Run” and get the ultra-realistic image of your description.
Result
Job Listings
- Applied Researcher, Sanctuary AI (Click here to apply)
Sanctuary AI is the leading AI company that develops control systems for humanoid robots. The company seeks applied researchers with 3+ years of experience implementing and deploying robotic manipulation tasks in simulation and physical robots. The candidate must have experience in ML and trained models in production and computer vision. The candidate must also have experience in Python, PyTorch, and TensorFlow. - Machine Learning Engineer, Sanctuary AI (Click here to apply)
Sanctuary AI is seeking a Machine Learning Engineer to join their team. The candidate must have experience of 3+ years in various ML methods with a focus on computer vision, natural language processing, etc. The candidate must be proficient in parallel computing and ROS and have experience working with cloud platforms.
This concludes our exploration of AI updates for this week. Stay tuned for more exciting updates next week! 🔥❤️⚡🤖📷
