Why You Should Pay Attention to Anthropic’s Rapid Growth
If you run a Malaysian SME, you probably do not spend your day tracking the revenue of AI companies. You are busy following up with customers, checking stock, approving payments, managing staff and keeping operations moving.
Still, a company’s rapid growth can signal where business software is heading. Anthropic, the company behind Claude, reportedly reached an annualised revenue run rate of US$65 billion at the end of July 2026, compared with US$47 billion in May and US$9 billion at the end of 2025. That pace suggests businesses are not merely experimenting with AI. They are beginning to use it in daily work.
For you, the important question is not whether Anthropic is valuable enough for a future public listing. The practical question is this: which repetitive tasks in your business should no longer depend entirely on manual effort?
TL;DR
Anthropic’s reported growth shows that companies are adopting AI tools for real business work, not just testing chatbots.
For Malaysian SMEs, the best starting point is a controlled workflow such as customer enquiries, quotation preparation, document handling or internal reporting—with human approval at important steps.
What This Means
Anthropic develops large language models, which are software systems trained to understand and generate text. Its Claude products can help users summarise information, draft replies, analyse documents, organise ideas and support certain business processes.
The reported US$65 billion figure is an annualised revenue run rate. This is not necessarily the company’s revenue already collected over a full year. Instead, it projects a full year based on recent performance. TechCrunch reported that investors expected Anthropic to finish 2026 at between US$100 billion and US$120 billion, although those figures are expectations rather than guaranteed results.
The same report said rival OpenAI had doubled its revenue to US$40 billion from US$20 billion at the end of 2025. The exact calculations may differ between companies, so you should not treat the figures as a direct comparison. The broader signal is clearer: demand for AI services is growing rapidly across the business sector.
For an SME, this does not mean you need to adopt every new AI product. It means AI is becoming a normal layer inside business software. It may appear in your customer relationship system, accounting platform, helpdesk, document management tool, inventory system or messaging workflow.
The useful question is not “Can AI do everything?” but “Which approved business task can AI help you complete faster and more consistently?”
How This Applies to Malaysian SMEs
1. Customer enquiries can be handled more consistently. Many Malaysian businesses receive enquiries through WhatsApp, Facebook, Instagram, email and website forms. A small team may answer the same questions repeatedly about operating hours, delivery areas, product specifications, appointment availability or required documents.
An AI-assisted workflow can classify incoming enquiries, suggest replies and send straightforward answers based on your approved information. For example, a Klang Valley cleaning service could use automation to identify whether a message is about residential cleaning, office cleaning or post-renovation work. The system could then collect the location, preferred date and service type before passing a complete request to a staff member.
You should still require human approval for unusual requests, complaints, refunds, sensitive personal information and anything that could create a contractual commitment. The benefit is not removing your team from customer service. It is allowing them to spend more time on cases that genuinely need judgement.
2. Quotations and proposals can start from structured information. Preparing quotations often involves copying customer details, selecting service items, checking terms and formatting the document. These steps are suitable for automation when your pricing rules and approval limits are clear.
Suppose you operate an electrical services company in Johor. A customer submits a request for wiring work. An automated process can extract the project type, location and requested date, place the information into a standard form and prepare a draft quotation for review. Your staff can then check technical details, availability and the final terms before sending it.
This approach is safer than allowing AI to invent prices or make technical promises. Use AI to organise information and prepare a draft. Keep commercial approval with an authorised person.
3. Documents and internal knowledge can become easier to use. Your business may already have standard operating procedures, staff guidelines, product catalogues, supplier documents and past proposals. The challenge is that these materials are often stored across folders, chat groups and individual laptops.
An AI search assistant can help staff find information from approved documents. A new employee could ask how to process a delivery issue or which documents are needed for a particular customer type. The answer should be based on your internal materials, with links or references so the employee can verify it.
This is especially useful if your business has a small management team and relies heavily on one experienced person. Capturing routine knowledge reduces interruptions and makes training more structured. Before using such a system, check how the provider stores data and restrict access to confidential documents.
4. Reports can be prepared without waiting for manual compilation. Owners often need simple answers: Which enquiries are still open? Which orders are delayed? Which customers have not received a follow-up? Which service category is creating the most support work?
An automated reporting workflow can collect information from your existing tools, organise it and produce a short summary for review. It can flag missing fields, overdue tasks or unusual changes. You still need reliable source data; AI cannot fix a process where staff record information inconsistently.
For a small distributor in Penang, a weekly report might combine sales orders, delivery updates and customer complaints. Instead of reading several spreadsheets, you receive a structured list of exceptions that need attention. That is a more practical use of AI than asking a general chatbot for broad business advice.
A Simple SME Adoption Framework
| Step | What You Do | Example |
|---|---|---|
| 1. Select | Choose one repetitive workflow | Replying to common enquiries |
| 2. Document | Write the current steps and rules | Approved answers and escalation conditions |
| 3. Control | Define what requires human approval | Refunds, contracts and complaints |
| 4. Test | Review a sample of outputs | Check accuracy before wider use |
| 5. Improve | Update information and remove errors | Refresh product details and procedures |
Do not begin with a broad objective such as “use AI across the company”. Begin with one workflow where you can measure whether response time, completion quality or follow-up discipline improves. Keep the scope small enough for you to understand what the system is doing.
Practical Takeaways
- List the repetitive tasks your staff perform every week.
- Choose a task involving clear rules and reliable information.
- Separate drafting from approval. AI may prepare an answer, but a person should approve important decisions.
- Create a single source of truth for product details, service terms and standard replies.
- Do not place confidential customer or employee information into a tool until you understand its data controls.
- Track simple measures such as response delays, incomplete enquiries and overdue follow-ups.
- Train staff to correct poor outputs instead of accepting every AI suggestion automatically.
- Review the workflow regularly as your products, policies and customer questions change.
The Bigger Picture
Anthropic’s reported growth is part of a wider movement in which AI capabilities are being built into ordinary business applications. You may not purchase a standalone AI system at all. Instead, your existing software may gradually gain features that draft messages, classify documents, summarise activity and recommend the next action.
This will raise the importance of process quality. Businesses with clear procedures, accurate records and defined approval rules will benefit more than businesses that simply switch on an AI feature. Automation needs structure. If your customer information is incomplete or your staff follow different procedures, the system will produce inconsistent results faster.
There will also be a stronger need for judgement. AI can help prepare information, but your reputation still depends on whether the answer is accurate, appropriate and aligned with what you promised the customer. Malaysian SMEs should be particularly careful with personal data, employment information, financial records and customer communications.
You do not need to follow every headline about AI funding, valuations or public listings. Watch for practical improvements that fit your operations. If a tool helps your team respond more reliably, reduces repeated administration and gives you better visibility over pending work, it may deserve a place in your workflow.
The most sensible starting point is a small, supervised use case connected to a real business problem. Once that process works, you can decide whether another workflow is suitable. That steady approach will help you gain useful automation without losing control of how your business serves customers.
Ready to Streamline Your Operations?
Your business should run itself. AutoRunBiz deploys AI agents to automate your daily operations — WhatsApp orders, invoicing, customer follow-ups, and accounting. Book a free 15-min ops audit to see where automation fits your business →