Morning All,

Anthropic quietly made a deliberate move that signals exactly where they think the future of professional AI sits. In this weeks edition we look at what that strategy really means, why AI in finance is not as far along as the hype suggests, and a three-step framework to actually get good at using these tools....

Anthropic Just Revealed Its Creative Industry Strategy

Anthropic quietly shipped 9 connectors that let's Claude directly control professional creative software, and the implications go well beyond a straightforward product update. Claude now operates inside common software products. Meaning it provides an additional intelligence layer for the tools creative professionals already use every day.

The connector list reads like a who's who of professional creative software: Adobe Creative Cloud (50+ apps including Photoshop, Premiere, and Illustrator), Blender, Autodesk Fusion, Ableton, Splice, Affinity by Canva, SketchUp, Resolume, and Claude Design. Furthermore, Anthropic became a Blender Development Fund patron at $280,000+ per year and is partnering with RISD, Ringling College, and Goldsmiths University on curriculum development around these tools. This is clearly a long-term positioning play.

The strategic contrast with OpenAI is genuinely fascinating. OpenAI went native...building creative capabilities directly into ChatGPT with image generation and Sora. Anthropic have gone in the opposite direction. Rather than replicating creative tools, Claude becomes the intelligence layer that works *within* them. Think of it like this: OpenAI built a new kitchen and invited you to use it to cook, whilst Anthropic hired world-class chefs and sent them to help you cook in your own kitchen.

Both approaches have real merit, but they serve different users. A filmmaker who lives in Premiere Pro doesn't want to leave their workflow to prompt an AI in a separate tab. They want intelligence embedded where the work actually happens. That's exactly what Anthropic is betting on.

These connectors serve professionals who already know Photoshop and Blender etc. The consumer creative market...face swaps, talking photos, style transfers, lip syncs...remains a huge opportunity. Platforms like Magic Hour, Canva's expanding AI features, and others are serving that layer entirely separately.

Anthropic just staked out a clear position in the creative industry, and it's one built on deepening existing professional workflows rather than disrupting them. They aren't the only ones benefitting from embedding agentic AI into their workflows. Financial services account for approximately 10% of global GDP...and most of the actual work is done digitally. It doesn't take a genius to see the incredible scope and impact agentic AI could have in this sector.

How Sage and AWS are Bringing Agentic AI to SMB Finance

Sage and AWS have expanded their partnership to embed agentic AI directly into SME finance workflows. and the timing matters because according to IDC, global AI spending is expected to grow by 31.9% annually between 2025 and 2029.

The core problem Sage and AWS are solving is accessibility. Cost and complexity have long been the barriers stopping SMEs from modernising their finance systems. Most small business owners don't have a CTO to navigate cloud migrations or AI integrations. They just need it to work.

To that end, the AWS Marketplace distribution model is actually quite clever. Sage partners and independent software vendors can build AI agents on AgentCore and list them for sale in the Marketplace, giving SME customers a straightforward channel to discover and deploy new capabilities without ripping out their existing systems.

Now SME's have access to a range of intelligent assistants that can handle specific tasks such as accounts payable, cash flow management, payroll processing, and compliance reporting. The kind of repetitive, time-consuming work that eats up hours every week.

Many SMEs are still running desktop financial software, which effectively locks them out of real-time insights and AI-powered tools. Sage and AWS are combining their respective strengths, product knowledge and cloud infrastructure, to make that transition faster, cheaper, and far less disruptive than it's traditionally been.

AI Is Quietly Rewriting the Rules of Finance

According to Intuit, 88% of financial services firms now use AI to support finance-related functions, and 59% of finance leaders are already leaning on it for everyday operations and decision-making. Finance teams have traditionally tended to look backwards, reporting on what already happened. AI flips that around, using historical data to model future scenarios, forecast cash flow needs, flag emerging risks, and spot market opportunities before they're obvious.

Fraud detection is one of the clearest wins. Advanced machine learning models continuously learn from the data they process, getting better over time at identifying anomalies that human analysts might miss. Think of it like having a colleague who reads every transaction report ever written and remembers all of it. Unlike a chatbot that answers a single question, agentic AI can plan and carry out multi-step tasks with minimal oversight. Things like end-to-end procure-to-pay workflows, invoice processing, and accounts payable can run largely on their own, with humans stepping in only to handle exceptions. As AI grows more adept at mimicking human reasoning, its value as a forecasting tool will only increase.

In this scenario, the concept of responsible AI becomes incredibly important. Fairness, transparency, and compliance aren't optional extras in finance. If there is no trust then the whole system breaks. “In finance, trust isn’t an abstract concept. Instead, it’s earned through precision, traceability and control. Systems must produce outputs that can be validated, explained and reconciled within strict audit and regulatory frameworks.” AI models learn from historical data, and consequently, they can inherit and amplify existing biases. In credit scoring and lending, that's not an abstract problem. It's a tangible risk that left unchecked, affects the quality of life and open opportunities available to millions of people.

The future is bright and the future is clearly agentic. But what about today? In the midst of all this hype about AI agents, are they currently having any meaningful impact in the way we work?

According to a survey of 321 U.S. finance and accounting decision-makers, the answer is no. The survey by The Harris Poll found, only 28% of the surveyed finance teams who have invested in AI are currently seeing a measurable financial impact. The impact is felt differently depending on the size of the company. AI implementation is further along in companies with at least 1,000 workers, but they were also more likely to have seen no measurable financial impact (72%, vs. 48% of smaller firms). The larger companies are also less optimistic that AI will begin delivering measurable financial results soon. 35% said it will take a year or more, compared to 15% of the smaller companies.

Smaller firms can be more nimble and move quicker with more flexibility. The scale of process mapping, data preparation and workflow optimisation needed in a smaller firm is also much less. There is much more scope for an IC in a SMB to make an outsized impact in their day to day life by efficiently utilising AI systems. As a result it isn't surprising there is much more optimism is smaller firms about the potential of AI. Building AI literacy now, even at a basic level, is one of the highest-return on time investments a finance (in fact any) professional can make today.

So what does that look like?

Three Steps to Actually Getting Good at AI

Most people are going about AI learning completely backwards. They're drowning in news, hoarding bookmarks of tools they've never opened, and watching tutorial videos instead of just...using the thing. Here's a practical three-step framework, to help you cut through a lot of the noise.

Step one - Carefully stay informed. The AI landscape moves fast enough that missing a week genuinely matters. GPT-5.5 and Claude Opus 4.7 dropping within the same week is a good example. If you don't know what's available, you can't use it. The catch? This step has an addiction problem baked right in. People start monitoring the news and end up spending entire days doom-scrolling AI Twitter, testing every shiny new tool, and worrying about what's coming next. Awareness is the entry point, not the destination.

Step two - Filter your inputs. If you try to follow Reddit, Twitter, LinkedIn, and a dozen newsletters simultaneously, you'll spend roughly 40 hours a week consuming mostly content you can't act on. Do you actually need to track every open-source model release? Probably not, unless you're running the kind of hardware that costs more than a car. Pick one or two curated sources, spend an hour a week, (including 5 mins reading this newsletter!) and move on.

Step three - Use the tools. This is the step where most people stall out, and it's the only step that actually compounds. Think of a task in your week that you resent doing...something real that's already annoying you. Hand it to an AI. It won't be perfect. You might have to go back and forth with it to get exactly what you want. But it'll be different, and probably faster than doing it manually, and that friction is where you develop genuine skill. Then do it again with something else. And again. And again...

You don't get better at riding a bike by reading about balance. You wobble, you correct, you build instinct. AI is the same.

If you remember nothing else: The gap between people who are good at AI and people who aren't isn't really about intelligence or technical knowledge. It's about reps. The biggest returns on time investments for creatives, and all knowledge workers is learning which of their workflows can reliably be transformed using AI systems. Find something in your workflow that quietly frustrates you every single week, give it to an AI system today, and see what happens. That one experiment will teach you more than a month of newsletters ever could... thanks for reading this far though, I truly appreciate it. Please come back next week ;-)