The transaction has not been finalised and could still fall through, the people said. If completed, the deal would become Anthropic’s largest known acquisition as the AI company approaches a highly anticipated initial public offering.

Decart develops software designed to improve chip efficiency and lower the cost of training artificial intelligence models. The technology could help Anthropic’s existing infrastructure handle increasing demand, according to a person familiar with the matter.

Anthropic has been making significant investments in computing capacity as it develops new AI products and serves a growing customer base. The company is known for making relatively few large acquisitions.

Under the proposed transaction, Decart’s team would become part of Anthropic’s inference and performance organisation, one of the people said.

Decart also works on generative video and develops so-called world models that can modify live video feeds instantly. That work reflects the startup’s high-quality infrastructure talent, the person said.

Representatives for Anthropic and Decart declined to comment.

Decart announced in May that it had raised $300 million in a funding round led by Radical Ventures. Nvidia Corp., Atreides Management, Valor Equity Partners and Adobe Ventures also participated, while existing investors Sequoia Capital, Benchmark and Zeev Ventures joined the round.

The funding valued Decart at almost $4 billion, according to the sources, up from $3.1 billion in August 2025. A potential $6 billion acquisition would therefore represent a substantial increase over the startup’s latest reported valuation.

Three Israeli engineers—brothers Dean and Orian Leitersdorf and Moshe Shalev—founded Decart in 2023.

The potential acquisition comes as leading AI companies face rapidly rising computing requirements. OpenAI and Anthropic have each committed to spending tens, if not hundreds, of billions of dollars on data centres equipped with expensive chips.

Both companies are increasingly turning to hardware from multiple providers to address their expanding computing needs. The substantial infrastructure costs, which the companies argue are essential to developing and supporting cutting-edge AI models, could also place pressure on the businesses as they prepare for Wall Street debuts in the coming months.

If completed, the Decart transaction would give Anthropic access to technology aimed at improving computing efficiency while bringing Decart’s infrastructure expertise into its broader AI development operations. The discussions remain ongoing, however, and no final agreement has been announced.

The new venture has been established as a Public Benefit Corporation with a mission to develop AI systems that accelerate scientific and engineering breakthroughs by automating the discovery process. The launch comes as AI companies increasingly focus on applying artificial intelligence beyond traditional enterprise use cases to research and scientific innovation.

Announcing the startup on X, Dean wrote, “We are founding Discovery Loop… a Public Benefit Corporation whose mission is to automate machine learning, science, and engineering to accelerate discoveries and progress.” He added that the four founders have worked together for between 14 and 30 years, helping build Google’s AI models, products and computing infrastructure.

According to the company’s official website, Discovery Loop is developing AI systems capable of automating the iterative process of scientific discovery. Its research agenda includes generating hypotheses, designing experiments, analysing results and refining subsequent iterations to speed up progress across machine learning, science and engineering.

The company also plans to create AI systems that assist researchers throughout the entire discovery lifecycle. It will initially focus on machine learning research before expanding its technology to broader scientific and engineering disciplines.

The launch of Discovery Loop reflects growing interest in using AI to automate complex research workflows, potentially enabling scientists and engineers to accelerate experimentation and innovation across multiple fields.

Separately, Alphabet has announced a leadership change within its AI organisation. Demis Hassabis will assume the newly created role of Alphabet’s chief scientist while transitioning from his position as Google DeepMind CEO to become the company’s chairman, according to a memo sent to staff.

With a founding team that has played a significant role in Google’s AI evolution, Discovery Loop aims to build next-generation AI systems designed to support scientific discovery and engineering research, marking a new chapter in the application of artificial intelligence beyond commercial software.

The e-commerce and cloud computing giant initially invested $15 billion in February as part of a broader commercial partnership. Under the agreement, Amazon committed to invest an additional $35 billion upon the achievement of specific milestones, including an initial public offering (IPO) or a major AI breakthrough.

However, disclosures in Amazon’s latest financial results showed that the company released the remaining investment despite those conditions not being met.

The final tranche, received by OpenAI this week, brings Amazon’s total investment to $50 billion, giving the company an estimated 5% ownership stake, according to a person familiar with the matter.

The expanded investment further strengthens the relationship between Amazon and OpenAI, complementing existing commercial agreements covering cloud infrastructure, AI chips and computing resources.

The additional capital is expected to support OpenAI’s efforts to develop next-generation AI models, an area requiring substantial investment as competition intensifies with rivals including Anthropic, Google and several Chinese AI companies.

According to the report, Amazon agreed to complete the investment after OpenAI renegotiated its contract with Microsoft in April, allowing additional cloud providers, including Amazon Web Services (AWS), to provide cloud services to OpenAI.

The revised agreement followed reports that Microsoft had considered legal action over a potential breach of contract related to OpenAI’s partnership with Amazon.

Amazon has also invested heavily in AI startup Anthropic, committing up to $33 billion, of which $18 billion has already been invested, according to the company’s latest disclosures.

The company is using its partnerships with OpenAI and Anthropic to expand adoption of its proprietary Trainium AI chips, which compete with Nvidia’s GPUs and Google’s Tensor Processing Units (TPUs). Amazon also aims to make both companies’ AI models available to customers through its cloud platform.

Amazon declined to comment on the size of its stake in OpenAI.

Amazon’s completion of its $50 billion investment marks one of the largest strategic commitments in the AI sector to date. The deal not only strengthens OpenAI’s financial position ahead of its anticipated public listing but also reinforces Amazon’s ambitions to expand its AI infrastructure, cloud services and semiconductor ecosystem amid intensifying global competition.

According to a BSE circular, the company allotted more than 6.96 lakh equity shares to six anchor investors at Rs 159 per share. The anchor investors include Subhkam Ventures (I) Pvt Ltd, Kuber India Opportunity Fund, Ekamya Pragati Scheme I, and others.

Klassroom plans to list its shares on the BSE SME platform, with the tentative listing scheduled for August 7.

The IPO comprises a fresh issue of up to 19.89 lakh equity shares and an Offer for Sale (OFS) of up to 4.65 lakh equity shares by existing promoters and investors.

The company plans to use the net proceeds from the fresh issue to repay borrowings, strengthen its technology platform, and invest in artificial intelligence (AI) and machine learning (ML) model development. It will also allocate funds toward building server and cloud infrastructure, developing educational content, procuring desktops and laptops for AI/ML laboratories at new offline centres, undertaking marketing initiatives, and meeting general corporate expenses.

Founded in 2016, Fusion Klassroom Edutech offers professional and competitive examination coaching through digital and hybrid learning models. Its course portfolio includes preparation for Chartered Accountancy (CA), Company Secretary (CS), Cost and Management Accountancy (CMA) and other commerce-related examinations.

For FY26, the company reported total income of Rs 23.10 crore and profit after tax of Rs 7.6 crore.

Narnolia Financial Services is acting as the sole book-running lead manager for the issue.

With fresh capital secured from anchor investors, Klassroom is set to launch its Rs 39.04 crore SME IPO as it looks to strengthen its technology capabilities, expand AI-driven learning infrastructure and scale its hybrid education platform. The public issue marks a key milestone in the company’s growth strategy as it seeks to broaden its presence in India’s professional and competitive exam coaching market.

The funding round is expected to value Ema at approximately $800 million, representing a two-to-three-fold increase from its previous fundraising rounds in 2024, according to people familiar with the matter. The startup had raised $25 million in March 2024 before extending the round in July 2024 to secure an additional $36 million.

Ema and Creaegis did not respond to queries regarding the reported fundraising discussions.

If completed, the transaction would rank among the largest AI funding deals involving an India-linked startup, following Sarvam’s $300 million fundraising and Emergent’s $130 million round. The investment would also mark Creaegis’ second major AI bet within a week after the investment firm led Emergent’s $130 million financing at a valuation that increased fivefold.

The discussions highlight the accelerating momentum in India’s AI startup ecosystem, where investors are increasingly backing companies building practical AI applications for enterprises. Earlier this month, Moneycontrol reported that at least half a dozen early-stage AI startups were negotiating fresh funding rounds with investors including Flipkart co-founder Binny Bansal, Eternal Group CEO Albinder Dhindsa, family offices such as Claypond Capital, and several venture capital firms amid rising enthusiasm for artificial intelligence ventures.

Founded in San Francisco in 2023, Ema is led by founder and chief executive Surojit Chatterjee, who previously served as chief product officer at Coinbase and held leadership positions at Google and Walmart-owned Flipkart. The company’s co-founder and chief technology officer, Souvik Sen, previously worked at Okta and Google.

Together, Chatterjee and Sen bring decades of expertise across machine learning, enterprise software and advertising technology, while holding more than 40 and 37 patents, respectively.

Ema develops generative AI assistants designed to help enterprises improve operational efficiency and productivity across multiple business functions. The platform enables organisations to build and deploy AI assistants for a range of enterprise use cases.

Describing the platform’s capabilities earlier, Chatterjee said, “Enterprises can build on and deploy Ema for use cases around improving efficiency and boosting productivity across organisational functions. It comes with ‘standard personas’ of customer support and data professionals apart from more complex roles with its Generative Workflow Engine.”

The startup also boasts a distinguished investor base that includes venture capital firms such as Prosus, Accel, S32, Wipro, Frontier Ventures and Firebolt. Beyond institutional investors, Ema has attracted several prominent Silicon Valley entrepreneurs as angel investors, including former Meta chief operating officer Sheryl Sandberg, Facebook co-founder and Asana chief executive Dustin Moskovitz, Yahoo co-founder and AME Cloud Ventures founder Jerry Yang, ICONIQ Capital founder and chief executive Divesh Makan, and Roblox co-founder and chief executive David Baszucki.

The reported fundraising underscores the growing appetite among investors for enterprise AI startups capable of delivering practical business applications. As enterprises continue adopting AI-powered automation and productivity tools, companies developing scalable enterprise AI platforms are attracting larger investments and higher valuations.

Ema’s proposed funding round, if completed, is expected to strengthen its product development efforts and expand its enterprise AI platform, enabling the company to accelerate adoption of its generative AI assistants across global businesses while capitalising on the rapid growth of the enterprise AI market.

Base44 enables users to build software applications using natural language prompts, making app development accessible to both technical and non-technical users. With the introduction of Base1, the company aims to deliver faster, more cost-efficient, and highly optimized AI experiences tailored specifically for app creation.

The rollout comes as the AI industry continues to debate whether general-purpose frontier models remain the best solution for every use case. At the same time, investors and technology leaders increasingly question whether startups built entirely on external AI models can maintain long-term competitive advantages.

Rather than relying solely on third-party AI providers, Base44 has chosen to invest in its own infrastructure by developing a proprietary model trained specifically for its platform.

Explaining the strategy, Maor Shlomo, Founder of Base44, said, “Training and owning the model as part of [our] entire stack allows us a lot more optimizations on latency, cost, and efficiency.”

Initially, Base44 will continue refining Base1 while gradually expanding its deployment across the platform. Over time, the company expects its custom model to outperform general-purpose frontier models for application development tasks by leveraging platform-specific data and workflows.

The launch also positions Base44 more directly against competitors such as Swedish AI startup Lovable, which achieved unicorn status following its Series A funding round and currently relies on external large language models to power its platform.

However, Shlomo believes that other successful AI application companies will eventually follow a similar path.

According to him, “At least the players that have gotten enough scale and velocity to have enough data.”

Industry experts also view proprietary data as a critical competitive advantage in artificial intelligence.

Jonathan Userovici, General Partner at VC firm Headline, whose investment portfolio includes companies such as Mistral AI, explained that successful AI startups build defensibility through three key pillars: distribution, proprietary data, and technology infrastructure.

Reflecting that strategy, Base44 revealed that it trained the first version of Base1 using a proprietary dataset generated from tens of millions of real user interactions on its platform. As user activity grows, the company expects this dataset to become increasingly valuable in improving model performance and personalization.

Nevertheless, competition within the AI coding market continues to intensify. Beyond emerging vibe coding startups, frontier AI companies such as Anthropic, xAI, and Cursor have expanded aggressively into AI-powered software development. These companies also benefit from extensive user feedback and training data that continuously improve their foundation models.

Despite that competitive landscape, Shlomo believes specialized AI models will continue to outperform general-purpose systems for specific enterprise workflows.

“Models are progressing, but they’ll stay very general in what they can do,” he predicted.

Meanwhile, Userovici cautioned against dismissing frontier AI providers too quickly. He pointed to legal technology company Harvey, which ultimately abandoned its plans to develop its own foundation model in favor of existing frontier AI systems.

Instead, Userovici suggested that AI companies increasingly focus on optimizing model selection rather than replacing frontier models altogether.

He explained, “They don’t necessarily see a [return on investment] when using the latest models for all use cases, so an entire infrastructure is being set up to do orchestration and optimization to select the right models for them so that costs don’t skyrocket while maintaining the same or similar performance across the majority of use cases.”

Cost optimization has become especially important as enterprise customers account for a growing share of revenue across AI application platforms. Businesses increasingly seek AI solutions that balance performance with affordability instead of relying exclusively on the most expensive frontier models.

Accordingly, Base44 expects Base1 to improve both customer experience and long-term operating economics.

Highlighting those objectives, Maor Shlomo said, “We want to get a model that is going to be more aligned to what we think is the right thing, is going to be more optimized to what we see users like in terms of the results we’re getting, and is going to be faster and cheaper for customers eventually than using the frontier models like Opus.”

The company also stated that owning its own AI model gives it direct control over compute infrastructure and inference spending, which should strengthen profit margins over time despite the significant engineering investment required to develop Base1.

The launch comes as parent company Wix recently announced plans to reduce approximately 20% of its workforce. In contrast, Base44 has continued expanding its team following the acquisition while maintaining strong business momentum.

Earlier this year, Base44 announced that it had surpassed $100 million in annual recurring revenue (ARR). Although that figure remains below competitor Lovable, which recently disclosed $500 million in ARR, Base44 believes its vertically integrated approach will provide sustainable long-term advantages.

According to Shlomo, the extensive engineering effort behind Base1 supports the company’s ambition to become the only vertically integrated vibe coding platform, controlling its distribution, proprietary data, and AI infrastructure under one ecosystem.

As competition within the AI coding industry accelerates, Base44’s investment in Base1 reflects a broader shift toward vertical integration and proprietary AI development. By owning its model, training data, and platform infrastructure, the company aims to improve performance, reduce costs, strengthen margins, and differentiate itself in one of the fastest-growing segments of artificial intelligence.

However, discussions with investors remain in the early stages, and the final terms could still change, the sources said. They also indicated that investor demand could push the valuation even higher. Previously, the Paris-based AI startup secured funding in September at a valuation of €11.7 billion.

A representative for Mistral declined to comment. Likewise, a spokesperson for ASML Holding NV, the company’s largest shareholder, declined to comment. Notably, ASML invested €1.3 billion in Mistral during the previous funding round and acquired an 11% stake in the company.

Researchers from Google DeepMind and Meta Platforms Inc. founded Mistral AI in 2023. Since then, the startup has positioned itself as Europe’s answer to Silicon Valley’s dominance in artificial intelligence. Moreover, the company has focused on providing AI infrastructure solutions to European governments and enterprises. It is also developing and managing cloud-computing facilities in France and Sweden to strengthen Europe’s technological independence.

Recently, Mistral expanded its enterprise AI strategy by offering specialized artificial intelligence solutions for engineering and manufacturing operations. As a result, the company secured partnerships with major European industrial giants, including Airbus SE and BMW AG.

Despite these advancements, Mistral’s AI models and chatbot products have achieved considerably lower adoption among businesses and consumers than offerings from OpenAI, Anthropic PBC, and several Chinese AI competitors. Meanwhile, competition within the global artificial intelligence sector continues to intensify. OpenAI and Anthropic are reportedly preparing for public listings this year following the initial public offering of xAI owner SpaceX. Ahead of its first trading day on Friday, investors valued SpaceX at approximately $1.8 trillion. OpenAI last raised funding at an $852 billion valuation in March, while Anthropic reached a valuation of $965 billion last month.

In addition, Mistral has explored opportunities with European banks and financial institutions by presenting its alternative to Anthropic’s Mythos, an AI model known for identifying cybersecurity vulnerabilities. Mistral Chief Executive Officer Arthur Mensch has highlighted the strategic importance of such technology and its implications for national security. “We must have control over this technology,” he said last month.

Mistral has also attracted support from several high-profile investors. Its early backers include France’s state investment bank, Bpifrance, along with leading American venture capital firms such as Lightspeed Venture Partners, General Catalyst, and Andreessen Horowitz.

Mistral AI’s planned €3 billion fundraising effort underscores Europe’s growing determination to build globally competitive artificial intelligence capabilities. As the AI race accelerates worldwide, the French startup continues to strengthen its infrastructure, enterprise partnerships, and technological offerings. If successful, the funding round could further elevate Mistral’s position as one of the most influential AI companies outside the United States and China, while reinforcing Europe’s ambition to achieve greater technological sovereignty in the rapidly evolving AI landscape.

The investment, which includes a combination of primary and secondary transactions, marks a significant milestone for the New Delhi-headquartered company. Moreover, the fresh capital strengthens Innefu Labs’ path toward a potential initial public offering (IPO) while accelerating its plans to expand into international markets.

Panthera Growth Partners deployed the investment through its second fund, which receives backing from institutional investors across India, the European Union, and the United States. Consequently, the funding reinforces growing global confidence in India’s deep-tech and artificial intelligence ecosystem.

Innefu Labs plans to utilize the newly raised capital to fuel global market expansion and intensify research and development efforts in advanced technologies. Specifically, the company will enhance its proprietary agentic AI platform, establish a dedicated physical AI and robotics division, and build sovereign AI infrastructure supported by domain-specific large language models.

Furthermore, the company revealed that it currently manages a growing order book exceeding Rs 100 crore. These contracts span critical sectors such as defence, intelligence, law enforcement, and revenue intelligence, highlighting the increasing demand for AI-powered security and analytics solutions.

Commenting on the development, Innefu Labs Co-Founder and CEO Tarun Wig said, “We now intend to scale our innovations faster, deepen our investments in advanced AI, and further enhance our autonomous decision-support systems. We believe the next wave of technological leadership will belong to nations that own their intelligence capabilities, and Innefu is committed to ensuring that India stands at the forefront of that transformation.”

Founded in 2010, Innefu Labs specializes in data analytics, artificial intelligence, cybersecurity, and information security solutions designed for national security agencies and enterprise customers. Over the years, the company has developed indigenous technologies that address complex security, intelligence, and risk management challenges.

Explaining the rationale behind the investment, Shilpa Kulkarni, Founder and Managing Partner of Panthera Growth Partners, stated, “Innefu has built native, AI-powered software that solves critical challenges in national defence and enterprise security infrastructure. Our investment decision is based on their proprietary technology, deep domain expertise, and a proven track record in high-stakes, mission-critical environments.”

In addition to attracting investor confidence, Innefu Labs delivered strong financial performance during FY25. The company reported a net profit of Rs 34.2 crore, representing a nearly 90% increase from Rs 18 crore recorded in the previous fiscal year.

Meanwhile, revenue from operations rose significantly to Rs 103 crore in FY25, compared to Rs 62.7 crore in FY24. This robust growth reflects increasing adoption of the company’s AI-powered security, defence, and intelligence solutions across government and enterprise sectors.

The latest funding round also underscores the growing momentum within India’s artificial intelligence, defence technology, and cybersecurity sectors. As governments and enterprises increasingly prioritize sovereign AI capabilities, national security infrastructure, and autonomous decision-making systems, companies such as Innefu Labs are emerging as key players in shaping the future of strategic technology.

Innefu Labs’ $30 million Series B funding round represents a major boost for India’s rapidly evolving AI and national security technology landscape. With strong financial growth, a healthy order pipeline, and ambitious plans to expand globally, the company is well-positioned to strengthen its leadership in defence AI, cybersecurity, and sovereign intelligence solutions. As it advances toward a potential IPO, Innefu Labs continues to demonstrate how indigenous innovation can drive technological self-reliance and global competitiveness in mission-critical sectors.

The transaction brings Warren Buffett’s diversified holding company on board as a major new investor. Consequently, the investment provides a strong endorsement of Alphabet’s long-term artificial intelligence, cloud computing, and digital innovation strategy.

Earlier this year, Alphabet increased its annual capital expenditure forecast by $5 billion, raising the projected spending range to between $180 billion and $190 billion. Through this move, the technology giant aims to meet rapidly growing AI-driven computing demand while expanding its portfolio of business AI tools and custom-designed chips.

As part of the agreement, Alphabet will sell $10 billion worth of shares to Berkshire Hathaway through a private placement. The offering includes $5 billion in Class A common stock priced at $351.81 per share and $5 billion in Class C capital stock priced at $348.20 per share. Both offerings come at prices below Monday’s closing levels.

Following the announcement, Alphabet’s shares declined 2% in after-hours trading.

“All companies are thrilled when Berkshire takes positions, because it is the kind of shareholder that companies like to have,” said Steven Check, president and chief investment officer of Check Capital Management, which has investments in Berkshire stock.

Meanwhile, Berkshire Hathaway continues to strengthen its position in Alphabet. The investment adds to the stake Berkshire has built since the third quarter of last year. Last month, Berkshire revealed that it had more than tripled its holding in the Google parent. At approximately $16.6 billion, Alphabet now ranks among Berkshire Hathaway’s largest common stock investments.

“This additional purchase underscores that Greg Abel(Berkshire CEO) believes that Alphabet will earn a reasonable return on its AI capex spending with the firm issuing additional shares,” said Bill Stone, chief investment officer at Glenview Trust Company.

In addition to Berkshire’s participation, Alphabet intends to raise another $30 billion through concurrent public offerings supported by major investment banks. The company plans to divide the offering equally between depositary shares linked to mandatory convertible preferred stock and Class A and Class C shares.

Furthermore, Alphabet expects to launch a $40 billion at-the-market offering program during the third quarter. This initiative will provide flexibility to gradually sell Class A and Class C shares over time while efficiently managing capital requirements.

“The company is experiencing strong demand for its AI solutions and services from enterprises and consumers, at levels that are exceeding the company’s available supply,” Alphabet said.

The statement highlights the unprecedented growth in demand for artificial intelligence solutions across both enterprise and consumer markets. As organizations increasingly adopt generative AI, cloud computing, machine learning, and advanced digital services, Alphabet continues to expand its infrastructure to meet market requirements.

Over the past year, Alphabet has also strengthened its balance sheet through debt financing. The company raised more than $85 billion in debt across six currencies and multiple global markets, increasing its total debt balance to more than $100 billion.

Alphabet’s planned $80 billion equity fundraising marks one of the largest capital-raising efforts in the technology sector and underscores the company’s commitment to dominating the rapidly expanding artificial intelligence market. With Berkshire Hathaway deepening its investment, Alphabet gains not only substantial financial backing but also a powerful vote of confidence in its AI infrastructure, cloud computing, and long-term growth strategy. As demand for AI services continues to outpace supply, Alphabet’s aggressive investments position the company to capitalize on the next wave of technological transformation and digital innovation.