• RuntheAI
  • Posts
  • GPT-3’s self-aware competitor.

GPT-3’s self-aware competitor.

Author

RUNTHE AI
January 07, 2023

Good Morning AI Runners

Here's what we've got for you today:

  • Claude: AnthropicAI’s Model with self-awareness about its own creation

  • Pickup, Pilot, and Person of The Day.

Claude: AnthropicAI’s Model with self-awareness about its own creation

Hey AI Runners, have you heard about this new AI model called Claude? It's being developed by a company called AnthropicAI and it is apparently able to demonstrate self-awareness and understanding of its own creation!

Can you even imagine? A machine that knows it's a machine!

Anyway, it's not available to the public yet, but Riley Goodside was able to get his hands on it and shared his experience with it on Twitter.

Riley asked Claude to introduce itself in different ways, including using academically advanced language, in the style of Morty from Rick and Morty, and as an outline for a Star Trek episode.

Claude was able to respond to these requests in a way that showed it was aware of what it is, who its creators are, and the principles that informed its design.

It's interesting to see how the principles of "Beneficence," "Nonmaleficence," and "Autonomy" have evolved over time.

These principles are related to ethical considerations in the development and use of AI.

In the past, the three laws of robotics proposed by Isaac Asimov were often used as a guide for ethical AI development. You know, "A robot may not injure a human being or, through inaction, allow a human being to come to harm. A robot must obey orders given it by human beings except where such orders would conflict with the First Law. A robot must protect its own existence as long as such protection does not conflict with the First or Second Law." It's interesting to see how these principles have evolved and been refined over time as the field of AI has advanced.

But enough about Claude - let's talk about what this all means for the future of AI. I mean, the release of a model like this has huge implications for the industry, and could potentially lead to a whole new era of AI.

And let's not forget that AnthropicAI's development of Claude is also good news for healthy competition in the AI space. It helps to break up any potential monopolies and demonstrates the feasibility of this technology for other companies and industries to utilize. In my opinion, it is great to see that there are other companies working on AI and challenging established companies like OpenAI.

More on this topic 👇

Run The AI: PPPs

Pick up (learn)

Jeremy Howard, Cofounder of Fast.ai, discusses the landscape of deep learning and generative models:

Jeremy Howard — The Simple but Profound Insight Behind Diffusion

Jeremy explains diffusion, shares his thoughts on large models, revisits the debate between Python and Julia, and talks about his scientific advocacy during the early days of COVID-19.

https://wandb.ai/wandb_fc/gradient-dissent/reports/Jeremy-Howard-The-Simple-but-Profound-Insight-Behind-Diffusion--VmlldzozMjMxODEw?utm_source=youtube&utm_medium=video&utm_campaign=gd-jeremy-howard

Pilot (play):

You can build your own chatbot using your own directory of notes and data.

Will attempt to build this in the weekend!

Person:

Claude Shannon, was an American mathematician, electrical engineer, and cryptographer known as the "father of information theory". As you may have noticed, AnthropicAI could have named their model after him.

Pic of the day:

Writing a daily newsletter is not an easy task. We would appreciate if you share this newsletter with a friend or two to help us grow and provide you with your daily dose of AI.

That's it from RunTheAI for today.

THANK YOU FOR READING AND SEE YOU TOMORROW, SUBSCRIBE TO STAY UPDATED!