GSV's AI News & Updates (05/10/24)
Mistral $6B Valuation, Alpha Fold 3, OpenAI and Publishers, Microsoft MAI-1, Take2 AI, Pearson AI Study Tools, Met Gala Deepfakes, Model Spec, India Election Content, OpenAI Search
General 🚀
OpenAI Is Readying a Search Product to Rival Google and Perplexity: the feature would let ChatGPT users search the web and cite sources in its results.
Mistral Set to Nearly Triple Valuation to $6B in Six Months: Mistral AI is close to raising roughly $600M from investors including General Catalyst and Lightspeed. Mistral, unlike some of the other LLM builders like OpenAI and Anthropic, is focused on taking an open source approach to its work. It’s one of the youngest of the LLM players in the market, and notable also for being a major effort to come out of Europe.
Google DeepMind releases Alpha Fold 3: AlphaFold 3 is a transformative AI model that predicts the structure and interactions of all biomolecules. This breakthrough extends beyond Alpha Fold 2's capabilities, encompassing DNA, RNA, and complex molecular assemblies, potentially revolutionizing fields from medicine to agriculture.
Leaked Deck Reveals How OpenAI Is Pitching Publisher Partnerships: the Preferred Publishers Program offers financial incentives to participating publishers, including guaranteed value (licensing payments for data access) and variable value (contingent on user engagement metrics).
Alibaba rolls out latest version of its large language model to meet robust AI demand: the latest model - Qwen2.5 - fares better than OpenAI’s GPT-4 model in language and creation capabilities, but fell short of its competitor in other categories like knowledge, reasoning and math.
Microsoft is building a new AI model, MAI-1, without OpenAI: the development of MAI-1 is being led by Mustafa Suleyman (former CEO of Inflection) and rumored to have ~500B parameters and is in a similar league to GPT-4.
ElevenLabs previews music-generating AI model: the company’s Voice Library currently allows users to sell their AI-cloned voice for money while maintaining control over its availability and how they’re compensated.
OpenAI working on new AI image detection tools: the tool will predict if an image is from DALL-E 3. They’re joining the popular C2PA standard for certifying content origins and adding metadata to everything created by their tools. (In 2023, OpenAI had to end a program that attempted to identify AI-written text because the AI text classifier consistently had low accuracy).
Meet My A.I. Friends: a NYT columnist spent the past month hanging out with 18 A.I. companions. They critiqued his clothes, chatted among themselves and hinted at a very different future.
Education and the Future of Work 📚
Take2 AI Raises $3M Seed Round to Transform Sales Recruiting Using AI-Powered Job Simulations: by leveraging generative AI, Take2 AI recreates company and role specific sales scenarios that not only immerse job applicants in the role, but also provide high-signal performance data to employers.
Pearson to Debut Generative AI Study Tools to University Students Internationally: AI tied to trusted Pearson content, personalized chat support, a simplified study experience, designed with privacy and ethical considerations in mind. “We know that students want to make their studying more efficient and are seeking ways to use trustworthy AI technology to get unstuck when they’re struggling in their coursework. We’ve built our AI study tools to meet this need, drawing from Pearson’s content to provide personalized support when students need it”.
Ava AI Chatbot Could Help Ease The School College Counseling Crisis: the College Guidance Network has developed Ava, an AI-driven chatbot that democratizes access to expert guidance and streamlines educational planning.
Community Colleges Are Rolling Out AI Programs—With a Boost from Big Tech: partnerships with tech companies, like the AI Incubator Network, are helping some colleges get the resources and funding they need to overhaul programs and create new AI-focused ones.
The Biggest Surprise from the Latest Survey of AI & Teachers: usage of generative AI continues to underperform expectations, usage of generative AI was highest among ELA / social studies teachers, teachers are reticent to use AI for fairly diverse reasons.
Harnessing the Power of Generative AI in Higher Education: A Call to Action for Educators: six areas on which to focus efforts and steer the integration of AI with intention and purpose.
The teens making friends with AI chatbots: teens are opening up to AI chatbots as a way to explore friendship. But sometimes, the AI’s advice can go too far.
AI Copilots Are Changing How Coding Is Taught: professors are shifting away from syntax and emphasizing higher-level skills.
YouTube's AI-powered 'Jump Ahead' feature rolling out widely to Premium users: uses AI to let you jump to parts of a video where most viewers typically skip ahead.
Startups and Tools 🛠️
7 Ways AI-Native Companies Can Improve Retention: what does retention mean for the huge wave of new consumer AI companies? 7 effective ways:
minimize the time it takes users to experience the magic of your product
feature-gate onboarding - let users hit a specific “aha moment”
design reciprocity
building smart notifications
gamifying products with streaks
give users summaries
status for power users
stick the landing - not every user may be worth retaining!
Docusign acquires AI-powered contract management firm Lexion: Lexion was incubated at the Allen Institute for Artificial Intelligence (AI2).
ScammerBlock: protect your parents and loved ones from phone scams.
Supertone Shift: real-time voice changer.
Pressmaster: AI-Powered PR Publishing.
Otto: your personal AI biographer.
Regulation and Policy ⚖️
OpenAI’s Model Spec outlines some basic rules for AI: the preliminary framework will guide the behavior of AI models, including GPT-4, emphasizing principles such as assisting users, benefiting humanity, and adhering to social norms and laws.
AI companies are making millions producing election content in India: Indian political parties are estimated to spend over $50 million on AI-generated election campaign material this year. Several young synthetic-media companies are working overtime to meet demand from all major political parties.
Microsoft bans US police departments from using enterprise AI tool for facial recognition: the changes in policy come a week after Axon, a maker of tech and weapons products for military and law enforcement, announced a new product that leverages OpenAI’s GPT-4 generative text model to summarize audio from body cameras.
Other
This year’s Met Gala theme is AI deepfakes:
For Conversations You Dread, Try a Chatbot: role-playing with an AI conversationalist can prepare you to handle difficult subjects with family, friends and colleagues.
Resources
The GSV BIG 10: weekly coverage corner for the top 10 stories, insights, and major plays in learning and skilling.
Thank you @Claire for the exciting AI insights!