Meta's Launching "Mini" AI Models That Could Save Your Business a Fortune
Brian Hanson
Digital Marketing & Real Estate Investment Enthusiast | Wealth & AI Speaker |
AI Smarter Than Smartest Humans By 2024?
Welcome to Another Issue of AI for Business
AI keeps on giving…and giving, like a bottomless gift bag!
Today, more game-changing and seriously cool updates from the biggest players will make you go ??…?? to ??.
Literally mixed feelings…
Excited?
Let’s get into it ??
Read time: 2.5 minutes
TRENDING AI NEWS
What happened: Meta is releasing smaller, cheaper versions of its upcoming Llama 3 large language model family as early as next week before the full Llama 3 model launches in the summer, which Meta claims to be “looser” than the previous version and will be able to answer controversial questions Llama 2 was not allowed to answer.
These smaller Llama 3 models aim to compete with recent entrants like Microsoft's 2B Phi and Google's Gemma in the efficient, task-specific AI assistant category.
??Fun fact: While these small models typically cannot handle long strings of instructions from users, they are faster, more flexible, and, most importantly, cheaper to run than a regular-sized model.
How it might impact you: While not ideal for long-form generation, the smaller Llama 3 models could enable businesses to run tailored AI assistants on-device for tasks like coding, API integrations, document summarization, etc.
The ability to self-host these relatively capable models privately could be a game changer for any enterprise, especially the smaller ones.
What happened: Google has released its most advanced generative AI model, Gemini 1.5 Pro, for public preview on its Vertex AI enterprise platform. Gemini 1.5 Pro has the ability to process between 128,000 and up to 1 million tokens of context at once - significantly more than competitors.
This massive context window will allow users to leverage the model for tasks requiring long context like analyzing large codebases, reasoning over lengthy documents, engaging in extended conversations, and cross-lingual content comparison.
How you can take advantage of it: For businesses grappling with large datasets, documents, or needing extended AI interactions, Gemini 1.5 Pro's context abilities could be helpful in opening opportunities to automate high-context professional workflows.
领英推荐
What happened: In a recent interview, Tesla CEO Elon Musk predicted that artificial general intelligence (AGI) that surpasses the smartest humans will likely arrive by next year or 2026 at the latest.
He also cited challenges like chip shortages currently constraining the training of xAI's Grok chatbot models. But said electricity supply may become the bigger bottleneck for scaling AI within 1-2 years.
Musk expects the Grok 2 model to be ready by May, requiring around 20,000 Nvidia H100 GPUs, while Grok 3 and beyond could need 100,000 such chips.
How it might impact you (not just you): If Musk's predictions prove accurate, the arrival of superintelligent AI in just a year or two could be enormously disruptive across every industry and domain.
Enterprises may need to brace for an acceleration in AI capabilities that could rapidly automate many white-collar professional tasks and cognitive workflows.
Does his view suggest that the competition to develop the first viable AGI system is intensifying? What do you think?
THE AI BUSINESS TOOLKIT
(Our Top 5 Weekly AI Picks)
Power Tools for Modern Businesses
WELL… THAT’S A WRAP!
Thanks for the support!
Join our?private AI group?on Facebook and explore more resources and get involved with our community ???
As always, I'd love to hear your thoughts, questions, or ideas. Don't hesitate to hit 'Reply' and share your insights with me!
Your friend,
Brian Hanson
CEO / Founder
This newsletter is free (for early subscribers). Enjoy!
Forwarded by a friend? Sign up with our exclusive newsletter here.