$695 Million Invested in 24 Hours
4IR - Daily AI News
Welcome back to 4IR. Here's today's lineup:
Lovable projects $1B ARR in 12 months with "vibe coding" revolution - when you don't need developers anymore
Cohere raises $500M at $6.8B valuation, Canada's answer to OpenAI - the non-American AI giant emerges
China's Darwin Monkey beats Intel's best neuromorphic computer - 2 billion neurons on 2,000 watts
Ex-Twitter CEO's stealth startup raises $30M for autonomous web agents - AI that browses the internet for you
🔥 TOP STORY: Lovable's "vibe coding" revolution: $1B ARR in sight, growing $8M monthly
The story: Lovable announced on August 14 it's on track to hit $1 billion in annual recurring revenue within 12 months, adding $8 million ARR monthly—possibly the fastest enterprise software growth ever recorded. CEO Anton Osika's "vibe coding" platform lets users generate complete applications from simple descriptions, positioning itself as the "Shopify for software." The startup closed a $200 million Series A at a $1.8 billion valuation, with enterprises abandoning traditional development for AI-first approaches.
What we know:
Growing at $8 million ARR per month
$1 billion ARR projected within 12 months
Valued at $1.8 billion after $200M Series A
"Vibe coding" eliminates traditional programming
Major enterprises replacing development teams
Fastest-growing enterprise AI company in history
Why it matters: This isn't just another coding assistant—it's the end of software development as we know it. If enterprises can build production apps without developers, what happens to the 27 million programmers worldwide? Lovable isn't selling tools to developers; it's selling their replacement to CEOs.
The $8M monthly growth tells the real story: companies aren't testing this, they're going all-in. When Fortune 500s start firing dev teams and pointing to Lovable, the entire $600B software services industry collapses. That's not disruption—that's extinction.
💰 FUNDING: Cohere's $500M round at $6.8B makes it the anti-OpenAI
The story: Toronto-based Cohere raised $500 million on August 14, hitting a $6.8 billion valuation with backing from Radical Ventures, Inovia Capital, NVIDIA, AMD Ventures, and Canada's Healthcare of Ontario Pension Plan. The company's cloud-agnostic large language models offer enterprises and governments an alternative to US-based AI providers, with total funding now at $1.5 billion. This positions Cohere as the leading non-American AI powerhouse amid growing concerns about AI sovereignty.
What we know:
$500 million raised at $6.8 billion valuation
Total funding reaches $1.5 billion
Strategic investors include NVIDIA and AMD
Canadian pension funds participating
Cloud-agnostic models for enterprise
Positioned as OpenAI/Anthropic alternative
Why it matters: While everyone obsesses over OpenAI vs Anthropic, Cohere quietly built the "Switzerland of AI"—neutral, enterprise-focused, and crucially, not American. For governments and companies worried about US data laws and export controls, Cohere isn't just an option—it's the only option.
The real tell? Both NVIDIA and AMD invested. When chip rivals back the same AI company, they're not betting on the technology—they're betting on geopolitics. Cohere just became the default choice for everyone who can't or won't use American AI.
🧠 BREAKTHROUGH: China's Darwin Monkey crushes Western neuromorphic computing
The story: Zhejiang University unveiled the Darwin Monkey neuromorphic computer in August 2025, achieving 2+ billion spiking neurons and 100+ billion synapses while consuming just 2,000 watts. The system's 960 Darwin 3 neuromorphic chips simulate macaque brain complexity, significantly exceeding Intel's Hala Point (1.15 billion neurons). This positions China as the neuromorphic computing leader despite semiconductor export restrictions.
What we know:
2+ billion spiking neurons (vs Intel's 1.15 billion)
100+ billion synapses
Only 2,000 watts power consumption
960 Darwin 3 neuromorphic chips
Simulates macaque monkey brain complexity
Developed despite export restrictions
Why it matters: Forget the chip wars—China just won the brain wars. While the West blocks advanced semiconductors, China built something better: computers that think like biological brains. This isn't about matching Western tech anymore; it's about making it obsolete.
The 2,000-watt power consumption is the killer detail. GPT-4 burns megawatts; Darwin Monkey runs on what your microwave uses. When you can simulate a primate brain on household power, AGI isn't a computing problem—it's an architecture problem China just solved.
🤖 AGENTS: Ex-Twitter CEO's $30M bet: AI agents that use the web for you
The story: Parag Agrawal, Twitter's former CEO, secured $30 million from Khosla Ventures on August 14 for his stealth startup "Parallel." The 25-person team is building infrastructure for AI agents to autonomously browse and interact with websites, addressing the critical gap in current AI systems that can't actually use the internet. This backend technology positions itself as essential infrastructure for the next generation of autonomous AI applications.
What we know:
$30 million from Khosla Ventures
Codenamed "Parallel" with 25-person team
Building web interaction infrastructure for AI
Enables autonomous browsing and task completion
Former Twitter CEO leading development
Targeting AI agent ecosystem infrastructure
Why it matters: Every AI company talks about agents, but none can actually use a website. Agrawal isn't building another chatbot—he's building the nervous system that lets AI interact with the 5 billion websites humans use daily. Without this, AI agents are just expensive autocomplete.
Khosla betting $30M on infrastructure nobody sees is the signal. When VCs stop funding models and start funding plumbing, the gold rush is over. The infrastructure wars just began.
🏢 ENTERPRISE: NTT DATA deploys AI across 75% of Fortune 100
The story: NTT DATA announced a comprehensive Google Cloud partnership on August 13, deploying Agentic AI across banking, insurance, manufacturing, retail, and healthcare for its clients—which include 75% of Fortune Global 100 companies. The partnership creates a dedicated Google Cloud Business Group with thousands of engineers, targeting 5,000 certifications. Early wins include Carrefour migrating 380+ workloads and NTT's Virtual Travel Concierge handling 3+ million monthly conversations.
What we know:
Serves 75% of Fortune Global 100 companies
Dedicated team of thousands of engineers
5,000 engineers to be certified
Carrefour: 380+ workloads migrated
Virtual concierge: 3+ million monthly conversations
Covers banking, insurance, manufacturing, retail, healthcare
Why it matters: This is how AI actually wins—not through press releases about AGI, but through boring enterprise deployments at massive scale. While startups pitch revolution, NTT DATA quietly put AI into three-quarters of the world's largest companies. That's not adoption; that's infrastructure capture.
The Carrefour numbers reveal the truth: 380 workloads migrated, 100 legacy apps consolidated. Companies aren't adding AI—they're replacing everything with it. NTT DATA just became the arms dealer for enterprise AI transformation.
💸 FUNDING FRENZY: $695M deployed in 24 hours across specialized AI startups
The story: August 14 saw an unprecedented funding velocity with $695+ million invested across specialized AI startups in a single day. Titan raised $74 million from General Catalyst for AI-powered IT services while acquiring RFA. MIT dropouts at Sola secured $21 million from a16z for back-office automation. Healthcare AI dominated with Reprieve Cardiovascular's $61 million for heart failure therapy and Citizen Health's $30 million for rare disease platforms.
What we know:
$695+ million total in single day
Titan: $74M for IT services AI (plus acquisition)
Sola: $21M from a16z for back-office automation
Reprieve: $61M for AI heart failure therapy
Citizen Health: $30M for rare disease AI
Focus shifting from LLMs to specialized applications
Why it matters: The AI funding game just changed—investors stopped chasing the next ChatGPT and started funding boring, profitable, specialized tools. When VCs throw $700M at niche applications in one day, they're not betting on technology anymore—they're betting on replacement economics.
Notice what's missing? No foundation models, no AGI promises, no "10,000x better than GPT-4" claims. Just AI that does specific jobs better than humans. The gold rush is over. The gold mining has begun.
⚡ QUICK HITS
Google Pixel 10 gets Tensor G5 "Gemini era" chip with on-device Camera Coach
Standard Pixel 10 adds telephoto lens, previously Pro-only feature
China's neuromorphic breakthrough: Darwin Monkey simulates primate brain on 2,000 watts
AI startups raise record $695M in single day, signaling sector maturation
Enterprise AI adoption paradox: Everyone's buying, nobody's deploying
Stack AI analysis: AI transition from experimental to essential infrastructure accelerating