Here are the 10 hottest AI startups of the year so far

2025/06/29 14:17

By Mark Haranas , CRN

Compiled by: Felix, PANews

These 10 AI startups are paving the way for the future of generative and agent-based AI around the world.

In 2025, AI startups will remain the hottest investment market for private equity and VC investors, with billions of dollars of funding continuing to flow into AI startups.

Many startups have become mainstream AI vendors in the areas of large language models (LLMs), customer service agents, and code generation.

Other AI startups, such as AI Squared, Morphos AI, and Writer, are focused on driving innovation by integrating AI directly into business applications, offering AI cost optimization products, or platforms for building agents.

IT research firm Gartner predicts that global spending on generative AI will reach $644 billion by 2025, a year-on-year increase of 75%.

Additionally, Gartner predicts that global AI service sales will reach $609 billion by 2028, driven by innovation around new generative AI capabilities and traditional AI technologies that are creating better predictive analytics and decision-making solutions.

CRN has selected the 10 hottest AI startups so far in 2025, which have created some of the most attractive AI innovations in AI agents, automation, knowledge graphs, accelerating AI applications, and solving complex business problems.

AI Squared

Executive: Darren Kimura, CEO

Headquarters: Washington, D.C.

AI Squared facilitates AI adoption by simplifying the integration of AI models into customer business applications, enabling teams to quickly deploy, experiment, and scale AI solutions.

AI Squared integrates AI through its SaaS and on-premises platforms, combining data sources with AI capabilities to embed intelligent insights directly into business applications.

The startup acquired reverse ETL platform Multiwoven last year to bolster its ability to simplify data and AI migration into applications, while also raising $14 million in funding.

Anthropic

Executive: Dario Amodei, CEO

Headquarters: San Francisco

Anthropic is one of the world's largest AI startups, currently valued at over $61 billion.

The startup unicorn is the owner of Claude, a popular language model that combines documents, tools, data, and web knowledge to solve complex problems and write code.

The AI ​​startup raised $3.5 billion in a Series E funding round in March, which it plans to use to advance the development of AI systems, expand computing power, deepen research on mechanism explainability and consistency, and accelerate its international expansion.

Anysphere

Executive: Michael Truell, CEO

Headquarters: San Francisco

Anysphere is a leader in the automated code industry and owns the popular AI code tool Cursor.

The startup has already won over big customers like OpenAI, Nvidia, Major League Baseball, and Uber, which use Cursor to analyze programmers’ actions and make code suggestions.

The AI ​​unicorn, which recently surpassed $500 million in annual revenue and is valued at around $10 billion thanks to new investments from Thrive Capital and Andreessen Horowitz, launched a new $200-per-month subscription plan for Cursor in June.

Cohere

Executive: Aidan Gomez, CEO

Headquarters: Toronto and San Francisco

Cohere provides innovative multilingual AI foundation models, search, and end-to-end AI products designed to solve real business problems.

Cohere's AI platform provides customers with security, data privacy, and optionality across all major cloud providers, private cloud environments, or on-premises deployments.

Last year, the startup raised $500 million from investors including tech giants Cisco Systems and AMD, bringing its total funding to nearly $1 billion.

Decagon

Executive: Jesse Zhang, CEO

Headquarters: San Francisco

Decagon provides AI-powered customer support agents designed to improve customer support by automating and resolving inquiries at scale.

The startup’s Agent, through its Agent Operating Procedures technology, enables customers to build, manage, and scale AI agents for chat, email, and phone to automate repetitive tasks, increase productivity, and empower support teams.

The startup raised $65 million in funding in 2024, led by Bain Capital, and plans to close a $100 million round later this year.

DevRev

Executive: Dheeraj Pandey, CEO

Headquarters: Palo Alto, California

DevRev provides an AI-native platform that integrates customer support and product development.

The startup allows customers to generate interconnected knowledge graphs to drive AI agents. Its AI products, Airdrop and Knowledge Graph, are designed to help customers go beyond automation by unifying data across all systems, turning complex processes into intuitive conversations that drive results.

The startup raised $100 million at a valuation of $1.1 billion in 2024. Pandey was previously the co-founder and CEO of Nutanix.

Morphos AI

Executive: Aram Chavez, Chairman

Headquarters: Tempe, Arizona

Startup Morphos AI is focused on helping general AI (GenAI) developers optimize their Large Language Model (LLM) Retrieval Augmented Generation (RAG) vector databases to improve search accuracy and reduce storage and energy costs.

One of the smallest startups on CRN’s list, Morphos AI offers a SaaS service that aims to make AI more efficient and cost-effective to deploy at scale by improving the way AI systems store and process information through its Green Vectors technology.

The company says that by seamlessly integrating with existing systems, it can reduce computing resources and costs associated with AI operations.

Perplexity

Executive: Aravind Srinivas, CEO

Headquarters: San Francisco

Perplexity uses AI models such as GPT-4 and Claude to understand user queries, search the internet in real time, and aggregate information. The company provides a free AI search engine and information discovery platform.

Perplexity recently launched Perplexity Labs, which makes reports and spreadsheets for dashboards and simple web applications possible through a suite of tools built around deep web browsing, code execution, and chart and image creation.

The startup recently partnered with Nidia to provide localized and autonomous AI models to European customers.

Thinking Machine Labs

Executive: Mira Murati, CEO

Headquarters: San Francisco

The youngest startup on the CRN list is Thinking Machine Labs, founded and led by Mirati, former CTO of OpenAI.

The startup said it is developing a wide range of AI systems, focusing on AI programming and building multimodal AI models with advanced reasoning capabilities.

OpenAI co-founder John Schulman is the startup’s chief scientist, while CTO Barret Zoph has played a key role in OpenAI’s breakthrough innovations. The startup is seeking $1 billion in funding even though it has no revenue yet.

Writer

Executive: May Habib, CEO

Headquarters: San Francisco

AI startup Writer offers an end-to-end agent-building platform that includes collaborative tools to build, activate, and supervise AI agents based on a company’s data and powered by Writer’s large language models.

From faster product launches to deeper financial research and better experimentation, companies are transforming their business processes with Writer.

The startup has won hundreds of clients over the past few years, including Accenture, Intuit and Marriott, and has attracted investors like Salesforce Ventures, Adobe Ventures and IBM.

Related reading: Overview of AI investment in the first half of 2025: 58% of global venture capital flows to AI

Disclaimer: The articles reposted on this site are sourced from public platforms and are provided for informational purposes only. They do not necessarily reflect the views of MEXC. All rights remain with the original authors. If you believe any content infringes on third-party rights, please contact service@mexc.com for removal. MEXC makes no guarantees regarding the accuracy, completeness, or timeliness of the content and is not responsible for any actions taken based on the information provided. The content does not constitute financial, legal, or other professional advice, nor should it be considered a recommendation or endorsement by MEXC.