- Superintelligence
- Posts
- Chipmaker vs. Nvidia
Chipmaker vs. Nvidia
Hey friends! We've got some exciting news to share: our logo has had a little makeover! While this isn't the final version, we’re crossing our fingers that we won’t need many changes for the next (and final) iteration. Oh, and you'll also notice a slight tweak to our name: we’re now Superintelligence AI! We hope you love the new look and name as much as we do. Stay tuned for more updates and, as always, thanks for your support!
The AI World Today
Free AI Revolutionizes U.S. Classrooms
US Startup Chipmaker vs. Nvidia
OpenAI Whistleblower Probe
$100M AI Fund for UAE Startups
Mapping Africa’s AI Startup Boom
Meet 100 GenAI Startups In India
+
AI Spotlight
Heads Up
Save the Date
Good Reads
Microsoft and Khan Academy Transform Teaching Tools

Screenshot: Khan Academy/Youtube
Khanmigo, an AI teaching assistant developed by Khan Academy and Microsoft, is now available for free to all U.S. teachers. This innovative tool aims to enhance teaching efficiency and effectiveness by reducing preparation time and supporting over 20 different teaching tasks. Khan Academy and Microsoft have partnered to offer Khanmigo at no cost to U.S. teachers, aiming to significantly reduce the time spent on lesson preparation, which often accounts for over 50% of teachers' workloads. Khanmigo assists with tasks such as generating assessment rubrics, developing quiz questions, designing lesson introductions, and supporting individual education plans. By making preparations more efficient, Khanmigo enhances teaching quality and student engagement. Khanmigo's initiative is part of a broader collaboration to explore the use of small language models like Microsoft’s Phi-3 to improve AI-based teaching tools, potentially allowing these tools to operate in areas with limited internet access.
AI Startup Tenstorrent Aims to Disrupt Nvidia's Reign

Image: Illustration of Jim Keller/Superintelligence AI
Jim Keller’s startup, Tenstorrent, targets high AI costs by designing efficient chips to compete with Nvidia. Tenstorrent's second-gen multipurpose AI chip boasts higher energy and processing efficiency, claiming 33% cost reduction compared to Nvidia's DGX. Tenstorrent's approach, avoiding high-bandwidth memory, lowers energy consumption and costs. Their innovative design includes independent CPUs within each core, adaptable for various applications, from smartphones to AI data centers. Established in 2016, led by renowned chip designer Jim Keller and 100 million raised capital, Tenstorrent aims to disrupt the AI chip market, focusing on flexibility and cost-effectiveness to cater to diverse AI needs.
Whistleblowers Demand SEC Action on OpenAI

AI Generated Illustration
OpenAI whistleblowers have filed a complaint with the SEC, urging an investigation into the company's restrictive non-disclosure agreements. The whistleblowers claim that OpenAI's NDAs force employees to waive federal whistleblower compensation rights and restrict disclosures to federal regulators without prior company consent. They also allege that OpenAI's agreements lack exemptions for reporting securities violations. US Senator Chuck Grassley emphasized the chilling effect these policies may have on whistleblower rights. OpenAI did not respond to requests for comment. The SEC has not confirmed the existence of a whistleblower submission.
Lear About Polynome's $100M AI Fund for Startups

Image: Illustration/ Superintelligence AI
Polynome Group, a UAE-based tech company, announced a $100 million fund to support AI startups starting Q1 2025. This initiative aims to boost technology, AI software applications, and robotics sectors. The announcement was made at the AI retreat by the Dubai Center for AI Applications and the National Program for Artificial Intelligence. The fund targets seed, Series A, and growth-stage startups, investing $500,000 to $5 million per company. It plans to support up to 40 companies over five years. Polynome Group is renowned for hosting the Middle East's largest AI summit, Machines Can See, and developing fundamental models and applied AI products.
Africa's AI Startups: Mapping the Future Landscape

Image: Screenshot/Afrilab
A new report, conducted by Briter Bridges and published with support from the Bill and Melinda Gates Foundation, maps the landscape of AI startups across Africa. The report highlights over 900 AI stakeholders, including 669 startups, emphasizing their roles in agriculture, healthcare, and finance. It identifies challenges such as limited funding and regulatory support while showcasing opportunities for growth in these sectors. The report provides detailed insights into the geographic distribution, funding stages, and technological focus of these startups, underlining the rapid evolution of Africa’s AI ecosystem. Africa's AI startup ecosystem is burgeoning, with significant contributions to various sectors. According to the report, 40% of AI startups focus on financial services, while 20% target healthcare. The ecosystem has seen a 40% increase in new AI startups over the past two years, reflecting the continent's growing tech innovation and global interest.
GenAI Investments Propel India's Tech Industry Forward

Screenshot: Inc42
A recent media report highlights the dynamic landscape of generative AI (GenAI) startups in India, emphasizing the sector's rapid growth and significant investments. The report tracks over 100 GenAI startups that have collectively raised more than $600 million since 2019. These startups, such as SarvamAI and Krutrim, are leading India's AI innovation by developing indigenous large language models (LLMs) and other AI-driven solutions. The report underscores the importance of this sector in transforming various industries and enhancing technological capabilities in the country. With the GenAI market projected to surpass $17 billion by 2030, India's AI ecosystem is poised for substantial global impact and growth.
AI Spotlight
Stability AI Unveils Stable Diffusion 3 Medium

Screenshot: Stability AI/YouTube
Stability AI has launched its most advanced text-to-image model, Stable Diffusion 3 Medium, designed for both consumer PCs and enterprise-tier GPUs. This model offers exceptional quality, photorealism, and resource efficiency, making it a new standard in text-to-image generation.
Key Takeaways: Stable Diffusion 3 Medium is a 2 billion parameter model with superior image quality and photorealism. It efficiently runs on standard consumer GPUs and enterprise hardware. Available under an open Community License, with commercial licenses available on request.
How to Start: Download the model weights. Sign up for a free three-day trial on Stable Assistant or try it via Discord.
Partnerships and Future Plans: Collaborations with NVIDIA and AMD enhance performance and compatibility. Future updates will continue to improve features and user experience based on feedback.
Background: Stability AI's Stable Diffusion models have set a benchmark in AI-generated art, known for their innovation and accessibility in the generative AI space.
Heads Up
AI Service: Learn How Hunar AI Is Transforming Frontline and Volume Hiring in India
AI Service : SKT's AI-Powered Pet Healthcare Service to Make North American Debut
AI Partnership: Mastercard and Partners Align With UAE AI Strategy With Inaugural AI Challenge
AI Ambition: Australia to Pursue AI, VR Pilots for Aged Care Reform
Save the Date
September 16: Deadline of the Call for Submission of Papers to 31st International Conference on Computational Linguistics (COLING 2025) will take place in Abu Dhabi, UAE, January 19-24 2025.
Good reads

Image: Illustration/Superintelligence AI
July 12: SPIQA: A Dataset for Multimodal Question Answering on Scientific Papers