Artificial intelligence (AI) is accelerating at a breathtaking pace. If the progress made over the last six months is indicative of the future, we’re in for a thrilling ride.
Large language models (LLMs) like ChatGPT have evolved significantly this year, transitioning from a fun place to ask questions to a sophisticated system capable of coding, web design, report analysis, and complex problem-solving. This rapid evolution has attracted billions in investments, transforming the competitive landscape for LLMs like OpenAI‘s ChatGPT, Google’s Bard, and more.
But a new contender has emerged: Anthropic‘s Claude 2, released to the public last week and garnered considerable attention within the AI community. The company raised over $1 billion in funding, with $450 million from Google alone, marking one of the most significant AI investments since Microsoft invested in OpenAI. Other major contributors include Zoom, Salesforce, and more.
Claude 2 can assist with various tasks, from coding and mathematical calculations to legal research. It can also generate creative content like poetry, scripts, and more, with the ability to engage in conversation about virtually any topic.
Anthropic, founded by former senior members of the OpenAI team, seeks to distinguish itself by focusing on the ethical aspects of AI. According to their official website, they are committed to creating a “steerable,” safe generative AI.
If you’re looking for a powerful and versatile LLM, Anthropic Claude 2 is worth checking out. Despite being the new kid on the block, it’s already making a big splash with the ability to handle tasks that require longer context. For example, it can summarize lengthy PDFs with the same accuracy and quality as its well-funded rivals, minus the word limit of ChatGPT.
As a testament to its functionality, I recently uploaded a 20-page sell-side analyst report to Claude 2 and tasked it with extracting key takeaways and a price target. It pulled out the relevant financial metrics, prompted additional questions, and crafted a succinct article based on additional prompting. I see now why LLM prompting will be its own profession.
Screenshot of Claude.ai on desktop
Claude’s context window is best in class at 75,000 words compared to ChatGPT’s 3,000 words. If you need to review extensive and challenging texts, Claude is your go-to.
And Claude is no slouch in the intelligence department. It scored 76.5% on the bar exam’s multiple-choice section and 90th percentile on the reading and writing portion of the GRE.
However, Claude 2 is not yet connected to the internet, unlike Bard and Bing. It’s trained on data until December 2022, so it can’t provide real-time information like current stock prices or updates on Meta’s new Twitter rival, Threads.
Don’t let this discourage you from trying out Claude 2. The next public release is likely to feature internet connectivity, and even now, you can still input recently published web content, and it’ll respond to relevant queries based on that text.
The race for dominance in the LLM market is still on, and Claude 2 is proving itself as a formidable contender. Already a vast improvement over its predecessor, Claude 2 stands toe-to-toe with Google’s Bard and OpenAI’s ChatGPT. Accessible to the public via a beta website and to businesses through an API, Claude 2 is poised to evolve and compete with other LLMs in the years ahead.
The landscape of AI tools currently offers various distinctive features and advantages. To make the most out of these tools, consider leveraging each one based on your specific needs and preferences. To delve deeper into the world of artificial intelligence, tap into the wealth of knowledge shared by our Chief Digital Asset Strategist, Nick Black, all available for free.
Director of Technology Investing Research, Money Map Press