#025a Saturday Feature - India’s AI Talent Pipeline Plans, Squeezing More Power Out Of AI Chips, Claude Pro: The New AI Chatbot That Beats ChatGPT Plus.
Fresh & Hot curated AI happenings in one snack. Never miss a byte 🍔
This snack byte will take approx 7 minutes to consume.
AI BYTE # 1 📢 - Squeezing More Power Out of AI Chips: How Companies Are Getting Creative
⭐ As demand for advanced computer chips that power AI continues to grow, companies are exploring new techniques to get the most out of their AI chips.
Some companies are conserving their use of AI chips or turning to software that can increase the performance of available chips and help reduce costs.
Training generative AI models can require tens of thousands of graphics chips, most of which are produced by chip maker Nvidia.
The company has said it is securing more supply to meet the high demand. However, even if companies manage to find the necessary computing power, it can be cost-prohibitive.
To address this challenge, some companies are training multiple models at once or using central processing units (CPUs) for inference, which requires less processing power than training. These CPUs can be tuned with open-source software tools to get more performance out of them.
Other companies are turning to third-party software to optimize their GPU performance.
For example, Run:ai installs its optimization software directly on its customers’ GPUs, allowing it to automatically put idle computing power to use for better processing efficiency.
OctoML offers a similar service on top of the cloud, renting out access to processing power from cloud providers that it speeds up on behalf of its customers.
VMware, a titan in the software industry, is also stepping up its game in the AI arena. They’re not just focusing on virtualization anymore - they’re aiming to help customers manage their use of multiple cloud platforms and data centers for AI processing.
They are tailoring their virtualization service specifically for AI inference and fine-tuning models. This means enterprises are more likely to tailor AI models with their own data rather than training them from scratch. It’s all about customization and efficiency.
VMware has also updated its partnership with Nvidia, giving joint customers access to new Nvidia servers with fewer GPUs. These servers are perfect for fine-tuning, so companies won’t get blocked from using large language models due to insufficient processing power. Plus, they’re offering a price that can compete with what cloud providers charge for GPU use per hour.
Also, cloud providers like Amazon, Microsoft, and Google are now offering access to the processing power companies need by renting it out, just like they do with computer servers. And they are not stopping there - they are also developing their own silicon accelerators to speed up chip processing and lower costs.
Take Amazon Web Services’ Inferentia2 accelerator, for example. It’s 50% cheaper than the company’s standard computing service for training AI models. That’s a massive cost saving for companies looking to train their AI models without breaking the bank.
These strategies and techniques demonstrate the innovative approaches companies are taking to maximize the power of their AI chips and meet their processing demands in a cost-effective manner.
AI BYTE # 2 📢 - India’s AI Talent Pipeline: A Look At The Initiatives By IBM, Adobe, Meta.
⭐ India is a country with a huge potential for AI innovation and development. However, to realise this potential, it needs a strong and skilled workforce that can leverage the latest technologies and tools.
This is where the role of public-private partnerships becomes crucial, as they can provide the necessary resources and guidance to upskill the students and educators in the country.
One of the prominent examples of such partnerships is the collaboration between CBSE and IBM, which introduced an AI curriculum for high school students in Grades XI & XII in 2020.
This initiative, which has been implemented in about 200 schools across 13 Indian states, aims to equip students with AI knowledge and skills to build AI models for real-life use cases.
The curriculum covers AI basics, history, and applications, along with skills like design thinking, data fluency, and critical thinking. It also emphasises ethical decision-making and bias awareness in AI.
Another example is the partnership between Meta (formerly Facebook) and India’s Ministry of Education and Ministry of Skill Development and Entrepreneurship.
This initiative aims to equip students, educators, and entrepreneurs with digital skills and training in areas like digital marketing, AR, VR, AI, and Digital Citizenship.
Meta will work with organizations like NIESBUD, AICTE, and CBSE to provide diploma courses, online programs, and workshops to foster digital literacy and innovation.
Adobe has also launched a creative and digital literacy initiative called Adobe Express, in collaboration with India’s Ministry of Education. This program will impact 20 million students & 500,000 educators in India by 2027.
Adobe Express is an AI-powered content creation app that will enable students to express their ideas and stories through various media formats. The program will also offer a curriculum that covers creativity, generative AI, design, animation, and emerging tech, aiming to empower students with future-ready skills.
Besides these initiatives, there are also other collaborations between tech giants and academic institutions in India to advance research and innovation in AI and cloud.
For instance, IBM has renewed its research collaboration with IIT Bombay and IISc Bangalore to focus on areas like natural language processing, generative AI, machine learning for time series data, fake news detection, hybrid cloud optimization, and sustainable computing.
This partnership is part of IBM’s AI Horizon Network, which aims to tap into the intellectual talent of students, faculty, and researchers to address global challenges.
These developments indicate that a lot of public-private partnerships have been helping to upskill Indian students in AI and cloud technologies. However, while all these initiatives are beneficial for the country and the ecosystem, the country also needs additional support in the form of direct investment in research and AI development.
Multinational companies who wish to have a strong presence in India need advice and encouragement from the government to invest heavily in AI research and development.
This will not only boost the economy but also create more opportunities for the students and educators who are being trained in these domains.
AI BYTE # 3 📢 - Claude Pro: The New AI Chatbot That Beats ChatGPT Plus in Context and Usage.
⭐ Are you looking for a better AI chatbot service that can handle longer and more complex prompts than ChatGPT Plus?
If so, you might want to check out Claude Pro, the new paid version of Anthropic’s highly proficient AI chatbot "Claude 2."
Claude Pro is a game-changer in the AI industry, setting new standards with its remarkable context capability. While ChatGPT Plus manages a context of 32K tokens, Claude Pro astounds with a massive 100K tokens, offering a more enriched user experience.
Claude Pro also promises at least 5 times more usage compared to the free version of Claude.
That feature is set to cater to the demands of power user. This premium offering is structured to reset every 8 hours, thereby making it easier for users to initiate new conversations without any hindrance, and ensuring a seamless interaction experience.
Claude Pro is designed to natively handle extensive conversations, even when they involve large attachments. The document advises users to “start new conversations for new topics,” a guideline aimed at optimizing Claude’s performance and avoiding unnecessary re-uploads of files, which in turn conserves the message limit and expedites response times.
Claude Pro is also powered by an innovative “Constitutional” training method that differs from OpenAI’s Reinforcement Learning Through Human Feedback.
Anthropic’s method provides the model with a “Constitution,” a set of general rules that guide the AI to favor good over bad interactions, making it able to self-improve without human interaction, detecting bad behaviors, and adapting its conduct.
Claude Pro is the result of Anthropic’s impressive research and development in the AI field. Anthropic recently raised over $400 million in a funding round led by Google. The company was founded by former OpenAI researchers.
In contrast, OpenAI is mainly backed by Microsoft and has reached a valuation of nearly $30 billion.
If you are interested in trying out Claude Pro, you can sign up for a subscription on Anthropic’s website. You will get access to early features, priority of computing power, and unlimited support from the Anthropic team. P.S - As of now, its only available in the US and UK.
