Invest in the Future of Information

Invest in the Future of Information

It’s hard to imagine now, but there was a time when search engines were a free-for-all. Yahoo reigned supreme, dominating with its broad, consumer-friendly approach. AltaVista, a specialized search engine, catered to power users with advanced indexing and niche features. But neither of them would define the future of search. That role fell to Google—a latecomer that took a deeply research-driven approach, building a fundamentally superior system that outperformed its competitors in efficiency, scale, and long-term reliability. It didn’t just compete; it changed the game entirely.

A similar dynamic is playing out in AI today. OpenAI, like Yahoo before it, has captured the mainstream, embedding itself into enterprise workflows, consumer apps, and billion-dollar partnerships. Meanwhile, DeepSeek, much like AltaVista, is carving out a niche by pushing the boundaries of what’s possible—leveraging open-source models and rapid iteration to appeal to developers and researchers looking for transparency and flexibility.

But if history is any guide, the real AI giant of the future may not be the most popular or the most aggressive—it may be the one that builds the best foundation. That’s Anthropic’s bet. Rather than chasing engagement or accessibility, it has focused on AI that is safer, more controllable, and deeply aligned with human values. Its flagship assistant, Claude, prioritizes trust and reliability in an era when AI’s influence is growing exponentially.

Yet, just as Google’s methodical rise was overshadowed for years by more visible competitors, Anthropic has received far less attention than OpenAI or even the fast-moving open-source movement. The question is whether history will repeat itself. In a field where first-mover advantage often defines market winners, Anthropic is betting that the true long-term leader in AI won’t be the fastest or the flashiest—it will be the one that businesses, regulators, and society trust the most.

If the past is any indication, the biggest disruptor isn’t always the first out of the gate—it’s the one that gets it right.

Traction

Anthropic has rapidly evolved from a research-driven startup into a serious AI contender, fueled by enterprise adoption, strategic partnerships, and well-funded product development.

2021–2022: Laying the Foundation

Founded by former OpenAI executives, including CEO Dario Amodei, Anthropic set out to build safer and more controllable AI systems. Unlike OpenAI’s shift to a for-profit structure, Anthropic positioned itself as a public benefit corporation, prioritizing ethics over pure commercial incentives. Early research focused on Constitutional AI, a framework designed to embed human-aligned principles directly into AI models.

By the end of 2022, the company had begun private testing of its AI assistant, Claude, with enterprise partners like Notion, Quora, and DuckDuckGo. These companies used Claude for writing, search, and conversational AI, helping refine its capabilities ahead of a public launch.

2023: Entering the Market

Anthropic introduced Claude to the public in March 2023, launching with a 100K-token context window—a major advantage over OpenAI’s ChatGPT at the time. This positioned Claude as an attractive option for businesses needing AI capable of handling large-scale documents and research. Adoption grew, particularly in legal, finance, and customer service, where AI-powered contract review, document summarization, and knowledge management became critical applications.

By mid-2023, Anthropic released Claude 2, improving reasoning, speed, and safety, which further expanded its enterprise customer base. The company also secured major partnerships. In September, SK Telecom integrated Claude into its AI-powered customer service platform. In November, Anthropic raised a $450 million Series C round, led by Google and Spark Capital, bringing its valuation to $4.1 billion.

2024: Scaling Up

Anthropic’s biggest leap forward came in March 2024 with the launch of Claude 3, introducing three AI tiers—Haiku, Sonnet, and Opus—offering businesses greater flexibility in balancing speed, cost, and performance. Claude 3 Opus outperformed GPT-4 in key reasoning and coding benchmarks, strengthening Anthropic’s reputation as a top-tier AI competitor.

At the same time, Amazon invested $4 billion, securing AWS as Anthropic’s primary cloud provider and granting access to Amazon’s Trainium and Inferentia chips to improve efficiency. By mid-year, Claude 3.5 Sonnet set new standards in AI accuracy, speed, and usability, pushing Anthropic’s projected annualized revenue to $850 million, driven by enterprise adoption and API integrations.

In May, Anthropic partnered with Menlo Ventures to launch a $100 million AI startup fund, investing in early-stage companies building applications powered by Claude.

2025 and Beyond: The Next Phase of Growth

With AI adoption accelerating, Anthropic is expected to release Claude 4, further improving safety, reasoning, and multi-modal capabilities. Industries such as healthcare, finance, and government are likely to deepen their reliance on Claude, favoring its secure and explainable AI. International expansion will also be a focus, leveraging its partnership with SK Telecom while navigating regulatory approvals in Europe and Asia.

As Anthropic’s valuation climbs past $20 billion, speculation around a potential IPO is growing. While no plans have been announced, its positioning as a safety-first AI provider makes it an appealing candidate for investors concerned about regulatory risks and long-term AI governance.

Business Model

Monetizing Claude AI:

Anthropic generates revenue through Claude.ai, its flagship AI assistant. Anthropic’s Claude subscription tiers are designed to offer varying levels of access to its AI models, catering to individual users, teams, and enterprises:

• Free Plan: Limited access to Claude 3.5 Sonnet for casual users exploring the AI assistant for tasks like writing, summarization, and Q&A.

• Claude Pro ($20/month): Identical to OpenAI’s ChatGPT Plus, offers users up to 5x the usage of free tiers and priority access.

• Claude Team ($30/user/month): Expands usage limits with added tools for collaboration and admin controls. Offers access to all three Claude 3 Models: Haiku, Sonnet and Opus.

Model Pricing:

For enterprises, Anthropic offers an API-based pricing model (pay-as-you-go) for its AI. Costs range from $0.25 to $15 per million words input and $1.25 to $75 per million words output, depending on the model used.

Competitive Differentiation - Ethical AI:

Anthropic’s unique approach focuses on embedding ethical principles into its models, ensuring outputs are helpful, harmless, and interpretable:

1. Built-In Safety with Constitutional AI:

   Claude follows clear ethical principles embedded directly into its training. This reduces harmful outputs and ensures responses are both helpful and transparent. For instance, when faced with inappropriate requests, Claude explains why it cannot respond, balancing safety and clarity.

2. Reduced Risks for Enterprises:

   Businesses in healthcare, legal, and finance benefit from Claude’s focus on accuracy and predictability. Its safety-first design minimizes issues like biased outputs or AI “hallucinations” (fabricating incorrect information).

3. Aligning with Enterprise Values:

   Claude helps companies meet regulatory standards and corporate governance goals by offering AI that is reliable, steerable, and trustworthy—critical for industries where reputation and compliance matter.

4. Competitive Trust Advantage:

   In an environment where only 29% of business leaders trust commercial AI, Anthropic’s focus on safety and transparency bridges the gap, offering enterprises a dependable AI partner.

Strategic Partnerships and Cloud Infrastructure

Anthropic has secured deep-pocketed investments and cloud partnerships with major tech firms.

• Amazon: A $4 billion investment made AWS Anthropic’s primary cloud provider, giving the company access to Amazon’s Trainium and Inferentia chips to enhance AI efficiency.

• Google: A $2 billion investment and cloud collaboration strengthen Anthropic’s AI research and computing capabilities.

• Salesforce Ventures and Zoom: Additional backing from enterprise software leaders signals growing demand for Claude’s AI models in business applications.

By integrating AI safety with enterprise scalability, Anthropic has built a business model centered on trust, performance, and long-term adoption, setting it apart from competitors focused primarily on consumer AI.

Opportunity

Anthropic is positioned at the intersection of two powerful AI trends: the rapid adoption of large language models in enterprise settings and the increasing demand for safe, transparent, and controllable AI. While many AI companies prioritize speed and revenue maximization, Anthropic has differentiated itself by focusing on AI alignment and trust, making it an attractive choice for businesses in highly regulated industries.

One of Anthropic’s biggest expansion opportunities lies in enterprise AI adoption. Companies in legal, finance, healthcare, and government sectors require AI solutions that are reliable, interpretable, and free from unpredictable behavior—a gap that many general-purpose AI models fail to fill. Claude’s 200K-token context window, significantly larger than many competitors, makes it particularly suited for document-heavy industries like law and compliance. Additionally, Claude’s constitutional AI framework offers a structured approach to reducing bias and ensuring AI-generated responses adhere to clear ethical standards.

Another major growth area is international expansion. While OpenAI and Google dominate U.S.-based AI adoption, Anthropic is making inroads into global markets, with its partnership with SK Telecom signaling a push into Asia. As AI regulations evolve, governments and multinational corporations may prefer a safety-first model over more commercially aggressive alternatives, giving Anthropic a competitive advantage.

Anthropic is also well-positioned in the developer ecosystem. Companies integrating AI into their products often seek a balance between performance, cost, and safety—a gap that many competitors fail to address.

By offering three Claude models—Haiku, Sonnet, and Opus—Anthropic gives businesses flexibility to scale AI usage based on their needs. Its partnerships with AWS and Google Cloud also ensure that developers have seamless access to enterprise-grade cloud infrastructure, making Claude a viable long-term AI solution.

With the generative AI market expected to reach $1.3 trillion by 2032, Anthropic’s strong positioning in ethical AI, enterprise applications, and global expansion gives it a clear runway for growth beyond its current capabilities.

Competition

Anthropic operates in a space dominated by OpenAI, Google DeepMind, Cohere, Meta, and Hugging Face, each offering competing AI models. While OpenAI’s ChatGPT has become the most well-known AI product globally, Anthropic has carved out a strong niche in AI safety and enterprise AI applications.

How Anthropic Stands Out

Anthropic differentiates itself through its measured, research-driven approach. While OpenAI and Google push forward with rapid iterations, Meta focuses on open-source accessibility, and DeepSeek accelerates AI innovation with open-source models, Anthropic is betting that the future of AI will be shaped by safety, transparency, and enterprise trust.

Its strategic partnerships with Amazon and Google, combined with its Constitutional AI framework, position it as a long-term player in AI—one designed for businesses and institutions that need AI to be explainable, steerable, and secure.

Valuation and Investor Growth

Anthropic’s latest Series E round in March 2024, led by Amazon, valued the company at $18.4 billion. This represents a sharp increase from its $4.1 billion valuation in May 2023, reflecting investor confidence in Claude’s ability to compete at the highest level of AI development.

The company has raised $9.76 billion in total funding, with major investors including Amazon ($4B), Google ($2B), Salesforce Ventures, and Zoom Video Communications. These strategic backers not only provide financial resources but also position Anthropic within critical AI infrastructure ecosystems.

Anthropic’s projected $850 million in annualized revenue by late 2024 represents a 23x revenue multiple, making it one of the most valuable private AI companies behind OpenAI. Given the continued expansion of AI adoption in enterprise settings, Anthropic could see its valuation exceed $20 billion in 2025, particularly as it scales international operations and expands its cloud partnerships.

Investor interest in AI companies remains high, and Anthropic’s positioning as a safety-first AI provider makes it an appealing long-term bet for institutions concerned about regulatory risks, ethical considerations, and enterprise AI trust. While speculation around a potential IPO is increasing, Anthropic has not announced any formal plans to go public. However, given its growth trajectory and enterprise focus, a future public offering could generate significant demand from institutional investors looking to gain exposure to AI without the risks associated with consumer-driven models like ChatGPT.

With the AI sector still in its early innings, Anthropic’s strategic differentiation in AI safety, enterprise usability, and measured scalability gives it strong investor appeal, both in private secondary markets and potential public listings in the years ahead.

Pros and Cons

How to Invest

Two private market opportunities for Anthropic are currently available—one on EquityZen and another on Hiive. Hiive is offering shares at $70 plus a 4% fee, while EquityZen lists them at $61.98 per share with a 5% fee. Neither investment carries interest.

Both opportunities, however, have significant barriers: Hiive requires a minimum investment of $250,000, while EquityZen mandates a net worth of at least $5 million, making them inaccessible to most investors.

Notably, both investments are structured as funds, meaning investors in Hiive’s Anthropic SPV, at a minimum, will have the opportunity to sell their units in about six months. This suggests more opportunities to buy Anthropic stock may emerge, making it a trend worth monitoring for interested investors.

As always, if you want us to clarify anything in this material, shoot us an email at hello@coldcapital.co, and we’ll respond as soon as we can.

Disclaimers Below

_{This material has been distributed solely for informational and educational purposes only and is not a solicitation or an offer to buy any security or to participate in any trading strategy. All material presented is compiled from sources believed to be reliable, but accuracy, adequacy, or completeness cannot be guaranteed, and Cold Capital makes no representation as to its accuracy, adequacy, or completeness.}

_{The information herein is based on Cold Capital’s beliefs, as well as certain assumptions regarding future events based on information available to Cold Capital on a formal and informal basis as of the date of this publication. The material may include projections or other forward-looking statements regarding future events, targets or expectations. Past performance of a company is no guarantee of future results. There is no guarantee that any opinions, forecasts, projections, risk assumptions, or commentary discussed herein will be realized. Actual experience may not reflect all of these opinions, forecasts, projections, risk assumptions, or commentary.}

_{Cold Capital shall have no responsibility for: (i) determining that any opinions, forecasts, projections, risk assumptions, or commentary discussed herein is suitable for any particular reader; (ii) monitoring whether any opinions, forecasts, projections, risk assumptions, or commentary discussed herein continues to be suitable for any reader; or (iii) tailoring any opinions, forecasts, projections, risk assumptions, or commentary discussed herein to any particular reader’s objectives, guidelines, or restrictions. Receipt of this material does not, by itself, imply that Cold Capital has an advisory agreement, oral or otherwise, with any reader.}