April 01, 2026 – AI Daily Recap

From OpenAI’s record-shattering $122 billion fundraise to Anthropic’s embarrassing source code leak, today’s AI news cycle is packed with stories that reveal just how fast — and how unpredictably — this industry moves. Here’s everything you need to know.

OpenAI Raises $3B from Retail Investors in Monster $122B Fundraise

OpenAI has closed the largest private fundraising round in history: a staggering $122 billion round that values the company at $852 billion. What makes this round especially notable is the $3 billion carved out for retail investors — a move that signals OpenAI is laying the groundwork for an eventual IPO. The round was led by Amazon, Nvidia, and SoftBank, three companies with deep strategic interest in the AI infrastructure buildout, including the Stargate data center project that OpenAI is pursuing with Oracle and SoftBank. With annualized revenue now surpassing $25 billion, OpenAI is no longer just an AI research lab — it’s a commercial juggernaut moving rapidly toward the public markets.

Salesforce Unveils AI-Heavy Makeover for Slack with 30 New Features

Salesforce CEO Marc Benioff took the stage in San Francisco to announce a sweeping 30-feature AI overhaul for Slack. The centerpiece is a dramatically upgraded Slackbot, now powered by the company’s Agentforce AI engine, which can summarize threads, draft replies, surface action items, and coordinate workflows across connected enterprise tools. This is part of Salesforce’s broader strategy to transform Slack from a messaging app into a full-fledged AI-powered enterprise workspace. The move is a direct challenge to Microsoft Teams and its Copilot integration, raising the stakes in the battle for the AI-native workplace platform.

Anthropic Is Having a Month: Source Code Leak and Internal File Exposure

It’s been a rough stretch for Anthropic, the AI safety–focused company behind Claude. Two significant accidental data exposures hit the company in quick succession. First, around 3,000 internal files were made publicly accessible on March 27, including a draft blog post about an unannounced new model. Then, on March 31, a packaging error in the release of Claude Code v2.1.88 accidentally bundled nearly 2,000 source code files — over 512,000 lines of code — exposing the inner architecture of one of its key products. Anthropic called it “a release packaging issue caused by human error, not a security breach,” but the timing is particularly awkward as the company is simultaneously engaged in a legal battle with the Department of Defense over defense contracts.

Mercor Hit by Supply Chain Attack via LiteLLM Compromise

AI recruiting platform Mercor confirmed it was affected by a supply chain attack targeting the popular open-source project LiteLLM, an API gateway used by thousands of companies to route requests across 100+ large language models. The attack vector was sophisticated: hackers first compromised the Trivy vulnerability scanner via a misconfigured GitHub Actions workflow, then used that access to steal the PyPI publishing token for LiteLLM and push two malicious package versions (1.82.7 and 1.82.8). The malware stole SSH keys, environment files, cloud credentials, and AI API keys. Extortion group Lapsus$ subsequently claimed it had accessed Mercor’s data, including Slack conversations and contractor video recordings. Mercor, which works with OpenAI and Anthropic to train models, says it was “one of thousands” of affected companies.

Depthfirst Raises $80M Series B for AI-Native Cybersecurity

Just 90 days after emerging from stealth with a $40 million Series A, applied AI lab depthfirst has closed an $80 million Series B led by Meritech Capital, bringing total funding to $120 million at a $580 million valuation. The company is building what it calls “General Security Intelligence” — an AI-native platform that builds deep context on a company’s code, infrastructure, and business logic to identify complex vulnerabilities and generate ready-to-merge fixes. Depthfirst also announced its first in-house security model, dfs-mini1, initially focused on securing cryptocurrency smart contracts. The rapid back-to-back funding rounds underscore investor appetite for AI-driven cybersecurity at a time when supply chain attacks like the LiteLLM compromise are making headlines.

Kestra Raises $25M to Become the Orchestration Standard for the AI Era

Paris-based open-source startup Kestra raised a $25 million Series A led by RTP Global, with continued backing from Alven, ISAI, and Axeleo. Kestra is an orchestration platform for data, AI, infrastructure, and business workflows, and has seen enterprise revenue grow 25x over the past 18 months, executing over two billion workflows in 2025 alone. The funds will go toward Kestra 2.0 — featuring a new distributed execution engine and native agentic orchestration — as well as a fully managed cloud offering and expansion across North America and Europe. With over 30,000 organizations and 26,000+ GitHub stars, Kestra is positioning itself as the coordination layer that enterprises need as their AI workflows become more complex.

The Big Picture

Today’s news paints a vivid picture of an AI industry that is simultaneously reaching staggering commercial scale and grappling with the security and operational challenges that come with it. OpenAI’s $852 billion valuation and Salesforce’s aggressive Slack AI overhaul show that the biggest players are betting enormous sums that AI-native products will define the next era of enterprise software. But the Anthropic leaks and the LiteLLM supply chain attack are stark reminders that the infrastructure underpinning this AI boom — from open-source dependencies to internal release pipelines — remains fragile. Meanwhile, the rapid funding rounds for companies like depthfirst and Kestra suggest that investors see enormous opportunity in the tooling and security layers that will be needed to make enterprise AI actually work at scale. The theme is clear: AI is growing up fast, and the growing pains are very real.

Sources

• TechCrunch — OpenAI, not yet public, raises $3B from retail investors in monster $122B fund raise
• TechCrunch — Salesforce announces an AI-heavy makeover for Slack, with 30 new features
• TechCrunch — Anthropic is having a month
• TechCrunch — Mercor says it was hit by cyberattack tied to compromise of open-source LiteLLM project
• BusinessWire — Applied AI Lab depthfirst Announces $80 Million in Series B Funding
• Kestra — We are the Orchestration Control Plane of the AI Era: Kestra raises $25M
• Axios — Exclusive: Workflow orchestration startup Kestra raises a $25M Series A