This week in AI felt like someone finally pulled a fire alarm. Prices cratered, a $40 billion investment reshuffled the entire industry, and the question “which AI company do I actually trust?” just got a lot more complicated. If you run a business between $1 million and $20 million, the moves your competitors make in the next 30 days — about which tools to use, what to stop paying for, and who to bet on — could create a gap that takes years to close.
This Week’s 5 Questions
- Has the AI price war finally made AI affordable for your business?
- If you’re in professional services, did AI just replace your most expensive research tool?
- Should you worry that the company behind ChatGPT is in federal court this week?
- When Google bets $40 billion on a single AI company, what does that tell you?
- Are restaurants, retailers, and local service businesses about to see the biggest shift in digital advertising in a decade?
1. Has the AI price war finally made AI affordable for your business?
On April 24, Chinese AI lab DeepSeek launched its V4 model — the largest open-source AI model ever released — and immediately slashed prices by 75% compared to previous rates. V4 Pro, with a trillion-plus parameters and a one-million-token context window, now runs at roughly $0.15 per million tokens on input. For context: six months ago, comparable performance would have cost your developer team five to ten times that amount. At the same time, Google released Gemini 3.1 Flash-Lite at just $0.25 per million input tokens — about one-eighth the price of premium-tier models — with 2.5 times faster response speeds than earlier versions.
The practical implication: if your business is using AI for anything repetitive and high-volume — generating product descriptions, summarizing customer emails, categorizing support tickets, drafting estimates — you may be significantly overpaying. The price of doing those tasks just dropped through the floor. This is not abstract; multiple major AI vendors (OpenAI, Anthropic, Google) are expected to cut enterprise pricing within the next 60 days in direct response to DeepSeek’s move.
The question to ask your team right now is simple: what are we currently paying for AI tools, and when did we last shop that price against what’s available today? The answer might surprise you. If you’re locked into a contract signed before April 2026, you likely have legitimate grounds to renegotiate — and AI vendors know it.
2. If you’re in professional services, did AI just replace your most expensive research tool?
This one is specifically for law firms, accounting practices, real estate offices, financial advisors, and consulting shops — any business that pays for specialized research subscriptions. On April 24, Perplexity launched Patents beta, which it’s billing as “the world’s first AI patent research agent.” It autonomously searches patent databases, surfaces prior art, and generates analysis in minutes — work that previously required expensive subscriptions to tools like PatSnap or Westlaw, plus hours of billable associate time.
Perplexity also released “Personal Computer” this week: an always-on AI agent that runs on a dedicated Mac mini, connects to your local files, emails, and applications, and works autonomously 24 hours a day with a full audit trail and a kill switch for any sensitive action. The company hit $500 million in annual recurring revenue this week — growing fivefold — and is now positioned directly against Microsoft Copilot and Salesforce for enterprise knowledge work. Meanwhile, OpenAI launched Workspace Agents with native integration into Salesforce, Google Drive, Slack, and Microsoft 365, and Anthropic’s Claude for Word add-in completed its Microsoft Office rollout.
For any professional services firm, the question this week is: which of your recurring SaaS subscriptions does an AI agent now cover? The ROI math is shifting fast. A $17-per-month Perplexity Pro subscription that can do what previously required a $300-per-month legal research tool is not an experiment — it’s an immediate cost reduction. Walk through your software stack, identify the tools your team uses primarily for research and information retrieval, and schedule a pilot comparison this month.
3. Should you worry that the company behind ChatGPT is in federal court this week?
The Elon Musk vs. OpenAI federal trial opened Monday, April 27, in Oakland. Musk is alleging he was misled into funding the company when it was a nonprofit, and is challenging whether its shift to a for-profit structure was legitimate. Meanwhile, OpenAI is preparing for an IPO at a reported $852 billion valuation — the trial outcome could directly affect both the timeline and structure of that offering. Separately this week, OpenAI’s exclusivity agreement with Microsoft was formally amended, meaning OpenAI’s models can now be hosted by Amazon and Google as well.
And it wasn’t only OpenAI having a complicated week: Anthropic’s Claude experienced a service outage at 9:39 AM Eastern on April 27, xAI has been dealing with persistent connection errors as demand overwhelms its Colossus supercomputer, and Perplexity is facing a class-action lawsuit alleging it secretly shared user chat data — including Incognito sessions — with Meta and Google.
None of this means you should stop using these tools. But it is a strong signal that single-threading your business on one AI provider is a risk you probably shouldn’t take. The businesses that end up in the worst position during AI disruption aren’t the ones that moved too slowly — they’re the ones that were dependent on one platform when it hit a rough patch. If your entire customer support, content production, or sales workflow runs through one AI tool, now is a good time to build a backup.
4. When Google bets $40 billion on a single AI company, what does that tell you?
On April 24, Google announced plans to invest up to $40 billion in Anthropic — $10 billion committed immediately with up to $30 billion contingent on performance milestones. It’s the largest single investment in AI history. That same day, xAI (the company behind Grok, backed by Elon Musk) closed a $20 billion Series E round with NVIDIA and Cisco as strategic investors. And Cohere, a Canadian AI company focused exclusively on enterprise and regulated industries, announced a $20 billion government-backed merger with Germany’s Aleph Alpha, supported by the Canadian and German governments and Schwarz Group (the parent company of Lidl).
This is what market consolidation looks like in real time. Three separate $20-billion-plus moves in a single news cycle. What’s happening is that the AI market is organizing around a small number of heavily capitalized platforms — and the companies positioning for enterprise customers right now are the ones that will dominate procurement conversations for the next five years. For a $1M–$20M business, this matters because the tools you standardize on in 2026 are likely to be the tools your staff knows, your workflows depend on, and your competitors are catching up to in 2028.
The Cohere/Aleph Alpha story is particularly notable if you do business in Europe or work in healthcare, defense, finance, or any sector with strict data residency requirements. That merger creates the first serious “sovereign AI” alternative to U.S. providers — backed by both governments — and it signals that data privacy and jurisdictional compliance are becoming purchasing requirements, not just preferences. If you haven’t thought about where your AI vendor’s servers are located and what law governs your data, this week is a good time to start.
5. Are restaurants, retailers, and local service businesses about to see the biggest shift in digital advertising in a decade?
Meta’s AI is quietly becoming one of the most powerful tools for any business that depends on reaching local customers. This week, Meta’s Muse Spark model — a natively multimodal reasoning AI built under the new Meta Superintelligence Labs — officially began powering Meta AI across all of Meta’s 3 billion-user app surface: Facebook, Instagram, WhatsApp, and the Ray-Ban Meta AI glasses. Meta is now projected to surpass Google in global digital advertising revenue in 2026 for the first time: $243 billion versus Google’s $239 billion, driven by a 24% year-over-year growth rate versus Google’s 12%.
For a restaurant, retail shop, or local home services business — HVAC, plumbing, landscaping, roofing — this is a direct signal about where to focus your ad budget. Meta’s AI-optimized targeting is outperforming Google’s display network in return on ad spend, and the gap is widening because Meta keeps feeding its models with first-party social and behavioral data. WhatsApp Business is also becoming a more powerful customer communication channel, with AI features embedded directly in the messaging layer.
The practical move for any locally-focused business this month: if you haven’t run a structured A/B test of Meta ads versus Google ads in the last six months, you’re working off stale data. The landscape has shifted. Many small businesses that built their customer acquisition on Google Search and Maps are finding that Meta’s AI-powered campaign tools are generating better leads at lower cost — particularly for service businesses with strong visual appeal (a beautifully finished deck, a spotless kitchen install, a before/after landscaping job). Pull last quarter’s numbers, look at cost per lead by channel, and make the comparison honestly.
What This All Means
This week in AI had a single underlying theme: concentration. Prices are concentrating downward, capital is concentrating into a handful of platforms, and the gap between businesses that are actively using AI and those that are not is concentrating wider. DeepSeek’s price cuts and Google’s $40 billion Anthropic bet aren’t stories about technology — they’re stories about leverage, and which businesses have it. The ones who do are the ones making deliberate, informed decisions right now about what they’re paying, who they’re trusting, and what they’re measuring.
Not sure how any of this applies to your specific business? That’s exactly what our free audit is designed to answer. In 30 minutes, we’ll map out where AI can realistically save you time, generate leads, or give you visibility you don’t currently have — no jargon, no pressure. Book your free audit →