Status
Available
Last Round Valuation
$50 Billion
Share Price
$50.00
Already have an account? Sign In
Anthropic is an AI safety and research company working to build reliable, interpretable, and steerable AI systems. Anthropic's first product is Claude, an AI assistant for tasks at scale. Its research interests span multiple areas including natural language, human feedback, scaling laws, reinforcement learning, code generation, and interpretability.
The exponential growth in computational power for AI training suggests significant advancements in AI capabilities, potentially surpassing human intelligence within the next decade. However, ensuring these powerful AI systems are beneficial, honest, and safe remains a challenge. Rapid AI progress risks societal disruption and the premature deployment of untrustworthy AI, leading to potentially catastrophic outcomes.
Anthropic addresses AI safety by building large-scale generative models and conducting safety research to align them with human values, focusing on creating AI systems that are helpful, honest, and harmless. Their approach intertwines capability enhancement with rigorous safety measures, ensuring models are steerable, interpretable, and reliable. By leveraging their empirical science background, they aim to solve current AI challenges while preparing for future existential risks, grounding their work in practical applications to ensure both immediate and long-term safety.
Constitutional training is a process for training a model to adhere to a “constitution” of desired behavior. The core Anthropic model has been finetuned with constitutional training with the goal of becoming helpful, honest, and harmless.
Constitutional AI addresses these challenges by employing AI-driven feedback for output evaluation, utilizing a foundational set of principles for judgment, hence the "Constitutional" designation. This framework directs the AI to adhere to normative behaviors outlined in its "constitution," aiming to mitigate toxic or discriminatory outputs, prevent assistance in unlawful or unethical activities, and ensure the creation of AI systems that are beneficial, truthful, and safe.
Anthropic leverages a proprietary “Constitutional AI” research process built to align foundation models, prevent hallucination, and curb bias. Their technique combines supervised learning and reinforcement learning to teach “values” to a model. Anthropic’s first product, Claude, is a generic foundation model competitive with Open AI’s GPT-4.
Anthropic currently has two models: Claude and Claude Instant.
Claude is Anthropic’s most powerful model, which excels at a wide range of tasks from sophisticated dialogue and creative content generation to detailed instruction.
Claude excels at: (1) Complex reasoning, (2) Creativity, (3) Thoughtful dialogue, (4) Coding, (5) Detailed content creation .
Claude Instant is Anthropic’s faster and cheaper model, which can handle a range of tasks including casual dialogue, text analysis, summarization, and document comprehension.
Claude Instant excels at: (1) Performance at low cost, (2) Reduced latency, (3) Lightweight dialogue, (4) Moderation, and (5) Classification
The combined context window for Claude’s input and output is about 200,000 tokens, which works out to roughly 150,000 words, equaling over 500 pages of material, depending on the type of content.
The 200,000-token feature doubles what Claude could handle before, and it’s a significantly higher limit than the 32,000-token ceiling of the priciest version of Open AI’s GPT-4.
According to Forbes, Anthropic has also apparently signed a $3 billion contract with Google Cloud.
According to The Information in January 2024, after covering the expenses for customer support and the servers required for its AI operations, Anthropic achieved a gross margin, which is the gross profit expressed as a percentage of revenue, ranging from 50% to 55% in December 2023.
Anthropic charges users on a usage-based pricing model. Every million tokens (approximately 750K words) processed by Claude Instant costs $0.80 for prompts and $2.40 for completions. For Claude 2.0 & Claude 2.1, the price increases to $8.00/million prompt tokens and $24.00/million completion tokens.
In October 2024, Microsoft’s GitHub Unit announced GitHub's widely popular Copilot tool—previously powered solely by OpenAI models—will now also integrate AI models from Anthropic and Google.
In November 2024, Anthropic announced a partnership with data analytics firm Palantir and Amazon Web Services (AWS) to offer U.S. intelligence and defense agencies access to its Claude family of AI models.
In November 2024, Amazon.com Inc. announced it is investing an additional $4 billion in Anthropic. This follows a $4 billion investment in Anthropic completed earlier in 2024.
In December 2024, Bloomberg reported that Anthropic was recently on pace to generate around $875 million in annual revenue.
Your access to private investment
Customers
Nulla malesuada justo velit, eget
Donec rhoncus sapien sit amet mauris consectetur convallis. Croin mollis eros vel orci.
Key Investors
Pellentesque tempus, ipsum id suscipit tristique, Aenean lacinia quis tortor sed, ornare Integer vel.
Size of Market
Mauris velit nisl, ultrices ut est quis, dapibus euismod tellus. Ut finibus ultricies sollicitudin. Vivamus a aliquet magna. Fusce lacinia urna in laoreet facilisis. Aliquam dignissim efficitur sapien eu convallis.
Market Position
Nam volutpat purus vel lacus pellentesque porttitor sed in orci. Vivamus porttitor sit amet tellus non sollicitudin. Suspendisse sit amet turpis nisi. In sem velit, interdum at augue id, eleifend tristique nunc, Maecenas scelerisque.
Industry
Mauris velit nisl
Primary Vertical
Artificial Intelligence
Employees
500
Mosaic Score
840
Money
760
Momentum
850
Management
760
Market
960
What is a Mosaic Score?
| Name | Work History | Title | Status |
|---|---|---|---|
| John Smith | Aenean, quam eu, diam dignissim | Chief Executive Officer | Current |
| John Smith | Aenean, quam eu, diam dignissim | Chief Executive Officer | Current |
| John Smith | Aenean, quam eu, diam dignissim | Chief Executive Officer | Current |
| John Smith | Aenean, quam eu, diam dignissim | Chief Executive Officer | Current |
Last Round
Sept. 2019, Series B1
Valuation Post-Money
$400.2M
Amount Raised
$2.5M
Total Funds Raised
$75.2M
Implied Valuation
$84M
| Date | Round | Amount | Valuation | Investors |
|---|---|---|---|---|
| MM/DD/YYYY | Growth Equity | $1M | $2.8B | Name goes here |
| MM/DD/YYYY | Growth Equity | $1M | $2.8B | Name goes here |
| MM/DD/YYYY | Growth Equity | $1M | $2.8B | Name goes here |
| MM/DD/YYYY | Growth Equity | $1M | $2.8B | Name goes here |
| Company | Valued | Revenue |
|---|---|---|
|
Coinbase San Francisco, CA |
$2.8B | $1.3B |
|
Coinbase San Francisco, CA |
$2.8B | $1.3B |
|
Coinbase San Francisco, CA |
$2.8B | $1.3B |
