The stories worth your attention this week

Most AI news is noise. Tool launches with inflated claims. Research with methodology you’d need a doctorate to interrogate. Announcements designed to move share prices, not help marketing teams work better.

This week had four stories that cleared the bar. Here they are, with the context stripped out and the practitioner implication front-loaded.

1. Canva acquires two AI companies in one day

Acquisition

Canva acquires Simtheory and Ortto

On April 9, Canva announced the simultaneous acquisition of Simtheory (an AI collaboration and agent management platform) and Ortto (a customer data and marketing automation company). Both were founded by the same team — Chris and Mike Sharkey — who will join Canva in leadership roles.

Canva ended 2025 at $4 billion in annualised revenue with 265 million users. This is a clear structural bet that marketers want to do their entire workflow — design, data, automation, content — inside one tool.

What it means for you: If you’re evaluating your martech stack, Canva is explicitly positioning to collapse several tools into one. Worth watching whether Ortto’s automation capabilities get integrated into Canva’s free tier or stay premium-only.

2. Meta launches Muse Spark — its first serious AI model

Launch

Meta debuts Muse Spark, built by Meta Superintelligence Labs

Meta launched Muse Spark on April 9 — its first major AI model since hiring Alexandr Wang from Scale AI. The model is small and fast by design, powers the Meta AI app and desktop site, with rollout coming to Facebook, Instagram, WhatsApp, Messenger, and Ray-Ban glasses.

What it means for you: The more interesting signal isn’t the model itself — it’s that Meta’s AI infrastructure is now deeply integrated with its ad delivery system via Andromeda. A better underlying AI model improves Meta’s Advantage+ targeting. If you run paid social on Meta, platform ad performance is directly downstream of how well Muse Spark performs.

3. The Semrush AI content study every marketer needs to read

Research

Human content is 8x more likely to rank #1 — Semrush, April 2026

Semrush analysed 42,000 blog pages across 20,000 keywords. Human-written content appears at position one 80% of the time. Purely AI-generated content appears there 9% of the time. The uncomfortable finding: 72% of SEO professionals believe AI content performs at least as well as human content. The data says otherwise.

What it means for you: If you’re running AI content without substantive human editing before publish, this is a direct warning. AI-drafted content with real human editing comes within 4% of fully human content performance. See our full breakdown: Does AI Content Rank on Google?

4. OpenAI crosses $25B in revenue

Revenue

OpenAI surpasses $25B ARR; Anthropic approaching $19B

OpenAI has surpassed $25 billion in annualised revenue and is reportedly taking early steps toward a public listing. Anthropic is approaching $19 billion ARR. OpenAI now serves over 900 million weekly users.

What it means for you: The consolidation of AI capabilities into a small number of dominant platforms is accelerating. Building workflows around two or three AI providers — not just one — is now sound risk management.

What to ignore this week

Meta’s AI video generation announcement from IAB Newfronts is interesting but in beta with limited access. Don’t restructure your paid social creative process around a feature most teams won’t touch for six months. File it, check back in Q3.

Frequently Asked Questions

Why did Canva acquire both Simtheory and Ortto at the same time?
Both companies were founded by the same leadership team — Chris and Mike Sharkey — so it was likely a single deal structured as two entity acquisitions. Together they give Canva the infrastructure to evolve from a design tool into a full marketing operating system.
What is Meta Muse Spark and how is it different from Llama 4?
Muse Spark is Meta’s first proprietary (non-open-source) large AI model, developed by Meta Superintelligence Labs. Llama 4 was Meta’s previous open-source flagship which underperformed against expectations. Muse Spark represents a rebuilt technical approach and shift toward retaining the model as proprietary infrastructure.
Is the Semrush AI content study reliable?
It’s substantial — 42,000 blog pages, 20,000 keywords — but relies on GPTZero to classify content as human or AI-generated. AI detection tools are imperfect. The directional finding is consistent with other research, so it’s credible even if the specific percentages should be treated as indicative rather than precise.
Should I be worried about building marketing workflows on OpenAI tools given their scale?
Scale cuts both ways. A company at $25B ARR has strong incentives to maintain reliable API access but also has pricing power a startup doesn’t. Build core workflows that are model-agnostic where possible so they could be moved to a different provider without rebuilding everything.
Where can I follow reliable AI marketing news without the noise?
Marketing Brew (strong on paid social and ad tech), MarketingProfs AI Update (weekly digest), and Search Engine Land (SEO and search). Be sceptical of roundup-style posts that don’t cite primary sources for their statistics.