AI Gold Rush Begins

Can't read or see images? View this email in a browser

The VCs are restless and there is no stopping them. This week, Mistral AI, a young startup based in Paris, secured an astounding $113 million in seed funding led by Lightspeed Ventures — this, despite it being only a month old and lacking a substantial product or team. This funding also surpassed the previous record set by Stability AI, which raised $101 million in its seed round. Other notable AI startups like Anthropic and Cohere raised $124 million and $40 million in series A. “I’m 30 and have raised $100 Mn+ for startups I’ve founded,” said Steph Mui, the founder and CEO of PIN, a community-based startup investing platform. But how? Mui said you just need a $250K angel round to get started, and four weeks later you can pivot to AI and raise a $100 million seed round.  Amid what may appear as mere VC humour, there lies a bitter reality within the investment landscape. The relentless hype surrounding AI has reached such heights that it now threatens to be a bubble on the verge of bursting, resulting in a downpour of monetary consequences. Many experts predict that this could be much worse than the tech bubble witnessed last year or even the infamous dot-com bubble. Another AI startup, Synthesia, recently attained a unicorn status after raising $90 million in series C funding. “While we weren’t actively looking for new investment, Accel and NVIDIA share our vision for transforming traditional video production into a digital workflow,” said Victor Riparbelli, co-founder and CEO of Synthesia.  Now, a new report by McKinsey is giving VCs all the more reason to celebrate. The 68-page report, released yesterday, shows that generative AI is set to add up to $4.4 trillion of value to the global economy annually. The report further stated that half of all work will be automated between 2030 and 2060.  It also said that generative AI has the potential to change the anatomy of work, augmenting the capabilities of individual workers by automating some of their individual activities. So, any solution or product that taps into these possibilities is here to stay, and startups have a huge role to play. But, unfortunately, the insatiable greed of VCs poses a significant risk of undermining the potential of AI. Read: VCs will destroy AI. Cringing on AI “ChatGPT is just the tip of the iceberg" is one of the most favourite posts on the internet, put out by so-called ‘AI influencers’ and ‘AI experts’ who, out of nowhere, nudge you into believing that you are falling behind in life.  Social media platforms like Twitter, Instagram and LinkedIn are filled with “cringe” content around AI. This is because people are just riding this bandwagon of AI-powered tools, calling ChatGPT the news of yesterday. Read on. Edtech Dancing to Generative AI Tunes  Indian edtech startups are going gaga over AI, particularly generative AI and LLMs. The likes of upGrad, EdZola, Cuemath, SkillUp Online, and not to forget BYJU’s, are all busy experimenting with generative AI, introducing new products and services into the market.  upGrad told AIM that it is working on building its own proprietary LLM. “Cutting-edge Generative AI models have proven to be invaluable assets in enhancing our educational offerings and providing valuable insights to our learners. However, we also have an eye towards the future, as we are actively exploring the possibility of building our own proprietary LLM,” said Mayank Kumar, co-founder and managing director at upGrad. While the adoption of AI is good for enhancing productivity and increasing the efficiency of the platform, there are other concerns around the elimination of jobs for teachers and the quality of learning among other things. See what the edtech experts have to say here. When LLMs Try to Act Smart  Currently, open-source models such as Alpaca, Vicuna, and GPT4All, can be fine-tuned with the output of close-sourced ChatGPT with the hope to build stronger models. However, the imitation process is only going to lead to flawed results.  According to research, ‘The False Promise of Imitating Proprietary LLM’, fine tuning a weaker model to improve its capabilities has minimal to no impact on the model, as the base data remains the same and it only alters the style of the model. In addition, through this method, the new models will also inherit the flaws and biases of the stronger model.  But, now there is a new study that shows that training on SOTA LLMs output can make models forgetful. Read more to find out. TAUSIF ALAM & AMIT RAJA NAIK Thursday, Jun 15, 2023 | Was this email forwarded to you? Sign up here DOWNLOAD OUR MOBILE APP Stay Connected info@analyticsindiamag.com © 2023 Analytics India Magazine

This email was sent by support@analyticsindiamag.com to alexvarboffin.abbb@blogger.com Not interested? Unsubscribe | Manage Preference | Update profile Analytics India Magazine | 280, 2nd floor, 5th Main, 15 A cross, Sector 6, HSR layout Bengaluru, Karnataka 560102