the health strategist . institute
institute for
health transformation and digital health,
data driven & AI powered
Joaquim Cardoso MSc.
Chief Research Officer (CRO), Chief Editor,
Chief Strategy Officer (CSO), and
Independent Senior Advisor
Key points
- Claude Pro is here, and it’s going to be a rather big GAI product. This could be a direct competitor of both ChatGPT Pro and Inflection’s Pi and as the midpoint between those two products, I consider it favorable.
Some of the functions are:
- Level up your productivity across a range of tasks
- Including summarizing research papers (think about that for a moment!)
- Iterating further on coding projects
- The biggest selling point is definately that you’ll get five times more usage with Claude Pro when compared to the free tier, which means you can send more messages in a shorter period of time.
Anthropic has uncovered a new approach to AI safety that shapes the outputs of AI systems according to a set of principles. The approach is called Constitutional AI (CAI) because it gives an AI system a set of principles (i.e., a “constitution”) against which it can evaluate its own outputs.
- For certain kinds of indepth research, Claude might be superior to ChatGPT.
- While ChatGPT Plus manages a context of 32K tokens, Claude Pro astounds with a massive 100K tokens, offering a more enriched user experience. That means Claude Pro can handle longer and more complex prompts than its rival.
DEEP DIVE
Anthropic’s Claude Pro is a Major Generative A.I. Product
Claude Pro costs $20 per month in the US or £18 per month in the UK.
AI Supremacy
MICHAEL SPENCER
08/09/2023
Hey Everyone,
I was reading the profile of the founders on Time’s AI 100, and it dawned on me how much progress they too have made in the Generative A.I. gold-rush that has been 2023 in its first nine months.
But it only keeps getting better:
Claude Pro is here, and it’s going to be a rather big GAI product. This could be a direct competitor of both ChatGPT Pro and Inflection’s Pi and as the midpoint between those two products, I consider it favorable.
The benefits of the Claude Pro plan are:
In a world of so many A.I. tools and subscriptions, some tools are better than others.
The company says:
Since launching in July, users tell us they’ve chosen Claude.ai as their day-to-day AI assistant for its longer context windows, faster outputs, complex reasoning capabilities, and more. Many also shared that they would value more file uploads and conversations over longer periods.
As of early September, 2023, with Claude Pro, subscribers can now gain 5x more usage of their latest model, Claude 2, for a monthly price of $20 (US) or £18 (UK). Of all the products that have launched and that I have tried in GAI, this is one of them I’m frankly most excited about.
- Level up your productivity across a range of tasks
- Including summarizing research papers (think about that for a moment!)
- Iterating further on coding projects
- The biggest selling point is definately that you’ll get five times more usage with Claude Pro when compared to the free tier, which means you can send more messages in a shorter period of time.
- Do you envision using such a tool on a weekly or daily basis?
- Poe (for desktop) and ChatGPT Plus are both $20/ a month
Anthropic first launched Claude in March, 2023 which it markets as a bot that’s “easier to converse with” and “less likely to produce harmful outputs.” Its emphasis on trust and safety really is in stark contrast to OpenAI’s gunslingers and world-tour approach.
Anthropic says that Claude Pro users can expect to send at least 100 messages to Claude 2 every eight hours, when the message limit resets.
Claude is a bit like Pi, without being overly annoying. If ChatGPT is like a geek and Pi is like a dove, I find Claude 2 is a healthy midpoint between these two LLMs and products, so far as chatbots may or may not be your thing.
Get a huge discount and let your company’s learning and development fund cover the expense:
How we treat the future of A.I. really matters, and not just becoming like Microsoft, but becoming a guiding beacon for trust and safety in Generative A.I., which could lead to some fairly dangerous applications. In my mind in this sense, Anthropic and OpenAI are really polar opposites.
If OpenAI is Ravenclaw, Anthropic has to be Gryffindor, and all of this is actually important.
Anthropic, which was founded in 2021, has carried out pioneering “mechanistic interpretability” research that aims to allow developers to carry out something analogous to a brain scan-to see what’s really going on inside an AI system… — Time AI profile.
When you have managed to raise $1.45 billion at a valuation in the single-digit billions and your flagship product is a Chatbot, it’s going to be fairly good, even as compute is still darn expensive.
Poe’s Chatbot platform ( Desktop) by the makers of Quora is a great alternative, if you want to sample different chatbots. It seems the $20 price point for now is a bit steep for most consumers, and should go down once compute gets cheaper too on the backend.
The “tens of thousands of GPUs” to train its models necessary for companies like OpenAI, Inflection, Anthropic and its peers runs into the hundreds of millions of dollars.
Poe lets you pay one subscription for unlimited access to all of the bots on its platform for $19.99 per month or $200 per year.
Anthropic and Quora’s Poe seem to be good partners. Anthropic also has to compete with the likes of Enterprise GAI solutions like OpenAI, Cohere, AI21 Labs and so many others for foundational models to mesh with the needs of potential enterprise clients. Anthropic really is in the thick of it, and competition is formidable on all sides.
The number of tokens Claude 2 can handle truly is impressive. Claude 2 can take up to 100,000 tokens in each prompt, meaning it can work over hundreds of pages of text, or even an entire book. It’s also available on Amazon Bedrock.
Responding to the Series C Funding
“We are thrilled that these leading investors and technology companies are supporting Anthropic’s mission: AI research and products that put safety at the frontier,” CEO Dario Amodei said in a statement. “The systems we are building are being designed to provide reliable AI services that can positively impact businesses and consumers now and in the future.”
Anthropic really is an OpenAI spin-off and it shows. Spark Capital’s hired Fraser Kelton, the former head of product at OpenAI, as a venture partner. Spark was an early investor in Anthropic and doubled down as the lead in the Series C round. Famously, SBF and Ashton Kutcher of Sound Ventures also poured crucial funding into Anthorpic. Sam Bankman-Fried’s FTX and Alameda made a $500 million investment in the startup very early on, which is pretty ironic if you think about.
Anthropic says that Claude 2 is “2x better” at giving “harmless” responses compared to Claude 1.3 on an internal evaluation.
What might companies like OpenAI, Anthropic, Inflection, Cohere and others become? As I cover Generative A.I. startups and sometimes at the intersection of venture capital, it’s becoming super interesting.
I think 2023 and 2024 will be the gold-rush proper for GAI (Generative A.I.) startups. This really is just the beginning.
All of this means you can trust Claude Pro a lot more, than most random A.I. tools. If X and LinkedIn prey on the hypetrain of A.I., Anthropic is already more low-key and grounded. I find Cohere also shares this humble quality. OpenAI is the opposite of humble or open, from all that I can tell. Unfortunately leaders at Microsoft, OpenAI and Google have been anything but humble bragging about A.I., including in Earnings calls of Meta, even as they bleed much of their best talent over “internal arguments”.
Solid research and humility does matter, and not preying into that small glimmer of consumer interest as when ChatGPT went viral (which likely was a Microsoft PR stunt anyways).
Claude Pro isn’t just our best bet to get helpful answers, it’s from a small startup that’s trying to do things differently. Yet they aren’t so small that they are irrelevant.
Anthropic is worth supporting, like I would an indie game developer I respect. Because they actually stand for something in a very complicated game of Generative A.I. helping us become more productive, creative and empowered.
Constitutional AI (CAI) shapes the outputs of AI systems according to a set of principles, with the goal of making a helpful, harmless, and honest AI assistant.
Anthropic has uncovered a new approach to AI safety that shapes the outputs of AI systems according to a set of principles. The approach is called Constitutional AI (CAI) because it gives an AI system a set of principles (i.e., a “constitution”) against which it can evaluate its own outputs.
For certain kinds of indepth research, Claude might be superior to ChatGPT.
While ChatGPT Plus manages a context of 32K tokens, Claude Pro astounds with a massive 100K tokens, offering a more enriched user experience. That means Claude Pro can handle longer and more complex prompts than its rival.
With Claude 2, Anthropic has built upon the chatbot’s existing capabilities with a number of improvements. will be a major test in its product-market fit.
Thanks for reading!
Originally published at https://aisupremacy.substack.com.
