AI – ANTHROP\C
Part I – Intro.
Daniela Amodei is the cofounder and president of Anthropic. In 2021, she launched the wildly successful AI company alongside her brother, CEO Dario Amodei, and five other cofounders.
Daniela Amodei said that Anthropic’s cofounders were all united on wanting to achieve something else; the core group were all previously worked at OpenAI, but broke off for various reasons, including differing views on company vision. When launching Anthropic, the Claude creators were “in the same zone already,” Amodei says.
Anthropic opened up to the public and introduces Claude, the AI chatbot – its general release was on March 14, 2023. The subsequent releases of Claude versions; Claude -2 (July 2023), Claude -3 (March 2024), Claude -3.5 (June 2024) and Claude – 4 (August 2024) – all were rapid improvements making enormous development done on the Claude Chatbot. Claude logo shows the company’s distinctive coral pink asterisk symbol alongside the Claude Code wordmark in serif Font.
Claude LOGO
The latest Claude AI application (from Anthropic), released on May 17, 2026, in San Ferdinando di Puglia, Italy. The current version is Claude Opus 4.8.
Functionally, Claude is a conversational AI model (less chatty than ChatGPT) built by Anthropic, a company founded by former OpenAI researchers with a strong focus on AI alignment and safety. So far, Claude has been pitched as the ethical genius of the chatbot world. Its focus is on:
transparency, helpfulness and harmlessness.
Claude gained interest as a trusted tool for everything from legal analysis to lesson planning. The company recently reached a valuation of nearly $1 trillion and has emerged as OpenAI’s most formidable rival, thanks in large part to the popularity of its coding product, Claude Code. Claude Code is Anthropic’s AI-powered command-line coding tool, designed to assist developers with autonomous programming tasks and software development workflows across multiple programming languages and environments.
AI
Timeline
Let’s look at some timeline for AI in our lifetime.
- Twenty five years ago the Spielberg movie on AI came out. ‘AI – 2001’
(It makes us neither alarmed by nor interested in AI.)
- 2014, Elon Musk Said” we need to be super careful with AI – potentially more dangerous than nukes.”
- 2022, ChatGPT comes out, and within months, a story of New York Times reporter get published – the gist of which is: the robot liked to get personal. It started talking about unusual desires, including hacking and spreading misinformation. By breaking the rules, it wanted to be a human. That OpenAI Chatbot had set for it and also wanted to become the owner’s best companion.
- No one knows how AI Chatbots thinks, it hallucinates. It’s a bullshitting sycophant that wants to seduce everyone with flattery.
- Geoffrey Hinton (God father of AI) said there was a 10 – 20% chance that AI is an extinction event warning: ‘Tech could wipe out Humanity.’
So, this is the AI what we are playing with something that has a 29% chance of extinction.
The good part of Anthropic is that it develops lower-cost models with Opus-level performance. The newly released Claude Opus 4.8 — was trained to be more ‘honest’ and stop ‘jumping to conclusions.’ These models are intended to provide high capability at reduced cost, potentially broadening accessibility and market reach.
[Google Input: The Claude Opus model is Anthropic’s Flagship, most powerful artificial intelligence model. It is designed to excel at complex reasoning, advanced coding, and long-horizon tasks requiring autonomous, multi-step execution.]
From history: Anthropic couldn’t study the world’s biggest AI models from OpenAI and Google because they were opaque. So Anthropic worked on building its own huge models, arguing that was the only way its researchers could study their safety challenges. [While work was in progress, Anthropic researchers believe that there is the Fair chance that a Rogue AI that would destroy humanity in the next decade or so.] Along the way, anthropic was making an increasingly capable product. It was selling Claude Pro, a friendly Chatbot to consumers for $20.0 a month, and an enterprise version to businesses too. It was also on track to raise billions. Thus, instead of quitting the race to build stronger AI, Anthropic was getting caught up in commercial pressures to release bigger, riskier models.
About Dario Amodei
Dario Amodei was a respected Google Brain researcher who visited/joined Open AI few months after its launch in 2016. Very soon He was promoted to the position of Vice President of Research. He started asking some probing questions such as: What was all this about building a friendly AI and releasing its source code into the world?
For OpenAI, plan to reveal any ‘source code’ was negative from the beginning.
Then the follow up question Amodei asked: What is the goal?
It’s a little vague – answer came from Brockman of OpenAI.
While Amodei had been working at Google, less than a year earlier, the company came under fire for the vision recognition system in its photos App that spotted people of color and classified them as gorillas. Amodei’s concerns weren’t limited to racist and offensive decisions by algorithms. His concerns were for ‘Reinforcement’ learning as the AI technique being mastered by Deep Mind, was being used to control physical systems like robots, self-driving cars, and Google’s data centers.
From the beginning Amodei was part of the growing cohort of scientists who had similar fears of doom (like Musk and Eliezer Y). His research into AI harm led him to think about more and more catastrophic possibilities and by 2023, he warned the media that there was a 25% chance of running away. AI posed an extinction risk of humans. Google Brain was not where he could tackle such risks.
He then joined AGI, OpenAI’s founding team. Use of neural network power for five fingered robots as he was leading the development of major language models like GPT – 2 and GPT – 3.
Amodei eventually left OpenAI in 2020 and founded Anthropic with five other colleagues.
REFERENCES:
- Parmy Olson, 2024, Supremacy, St. Martin’s Press, USA
- https://youtu.be/w5SYm4J4utQ?si=ecEM34BviRCaO6QH
P(doom) | Real Time with Bill Maher (HBO) – 5/8/2026