🎙️ Webinar on Tuesday 18 November, 11am (Paris): Claude Duet vs OpenAI Dots, meet your new AI right hand

★★★★★

4,9/5 sur Google

📘 Corporate AI training: our free roadmap for a successful rollout

Download our AI roadmap

★★★★★

4,9/5 sur Google

Book a meeting

2026-10-25

What's new in Claude in 2026: Opus 4.6, Sonnet 4.6, Cowork… the complete guide to Anthropic's latest updates

Emma Pariente

Looking for the latest Claude update you shouldn't miss?
Anthropic has been releasing announcement after announcement since early February 2026. Opus 4.6, Sonnet 4.6, Cowork, Claude Code Security… in just a few weeks, the Claude ecosystem has moved to a completely different scale.

We review each new Claude feature, what it changes for you in practice, and why companies are switching in large numbers.

Since then, Anthropic has also launched Claude Opus 4.8 on 28 May 2026: we explain what changes in a dedicated article.

Here is everything you need to know 👇

‍

What's new in Claude: what changed in February 2026?
‍

In the space of three weeks, Anthropic rolled out:
‍

🚀 Claude Opus 4.6 (5 February) — the most advanced model ever released, with agent team capabilities and use inside PowerPoint
‍

🚀 Claude Sonnet 4.6 (17 February) — performance close to Opus… at Sonnet pricing ($3/M input tokens, $15 output)
‍

🚀 Claude Code Security (20 February) — a tool that scans codebases to identify security flaws
‍

🚀 Claude Cowork (major update on 24 February) — the AI agent platform for companies, with plugins, connectors and a private marketplace
‍

It is one of the densest sequences of updates ever seen from an AI provider!

‍

Also worth reading

Discover our other articles to go further on ChatGPT and its new features:

👉 ChatGPT Go: price (€8) and differences vs ChatGPT Plus 👉 ChatGPT App Store: 100+ apps (Canva, Spotify, Adobe, Booking, etc...) 👉 Shopping Research: ChatGPT's new AI shopping assistant

‍

New Claude feature no. 1: Opus 4.6, the smartest model on the market

‍

Release date: 5 February 2026

‍

Claude Opus 4.6 is Anthropic's flagship model. It is currently considered the best-performing model in the world for code, autonomous agents and computer use.
‍

What sets it apart:

‍

✔️ It can carry out autonomous tasks for more than 14 hours (according to measurements by METR, an independent evaluation organisation)

‍

✔️ It excels at complex multi-step tasks: code analysis, refactoring, fixing bugs across several systems

‍

✔️ It integrates natively with PowerPoint, Excel and Chrome

‍

✔️ Its price has been lowered: $5 input / $25 output per million tokens

‍

In short: raw power, with a level of reasoning that surpasses anything that existed until now.

‍

Discover our AI training courses & book a call

Explore our different training modules and choose, with us, the one that best meets your AI needs.

📚 See our AI training 📞 Book a discovery call

‍

New Claude feature no. 2: Sonnet 4.6, premium performance at the standard price

‍

Release date: 17 February 2026

This may be the most strategic new Claude release.
‍

Sonnet 4.6 reaches a level of performance close to Opus 4.5… while keeping Sonnet pricing: $3 input, $15 output per million tokens.

‍

In practice, tasks that used to require an extremely expensive model become accessible:

‍

🔥 Analyse a 200-page annual report and produce a structured summary

‍

🔥 Refactor a complex codebase by identifying the dependencies

‍

🔥 Build a multi-step action plan from a strategic brief

‍

🔥 Compare 5 supplier contracts and highlight the key differences

‍

Book your 30-minute call now

A no-obligation conversation to build your AI training around Claude.

👉 Tell us about your AI project 👈

‍

1 million tokens: the memory that changes everything

‍

Sonnet 4.6 introduces a one-million-token context window (in beta).
That is twice as much as the previous version.

‍

To give you an idea of scale, in a single conversation you can load:

‍

📄 A 300-page contract and ask questions about it without losing anything

‍

📄 A complete legal file with its appendices

‍

📄 Three months of financial reporting in one go

‍

📄 An entire code repository for a technical audit

‍

No more splitting documents into chunks.
No more copying and pasting 15 times.
You load everything, ask your question, and Claude keeps the context from start to finish!

‍

Computer use: Claude carries out tasks on your computer

‍

On the OSWorld benchmark, Claude goes from 14.9% to 72.5%. A spectacular leap.

‍

What this means in practice:

‍

✔️ It navigates an Excel spreadsheet, sorts a column and applies a formula

‍

✔️ It fills in a multi-step web form

‍

✔️ It moves from one tab to another, copies data here and pastes it there

‍

No technical connector.
No complex integration.
It sees the screen and it acts!

‍

Book your 30-minute call now

A no-obligation conversation to build your AI training around Claude.

👉 Tell us about your AI project 👈

‍

New Claude feature no. 3: Cowork, the AI agent that settles into the company

‍

Launch: January 2026 (research preview) → Major update: 24 February 2026

This is probably the most disruptive new Claude feature for the professional world.

Claude Cowork turns Claude into a genuine agent platform for companies.
It is no longer a chatbot: it is a virtual colleague that acts within your business tools.

‍

What has just changed:

‍

🔌 New connectors: Google Drive, Gmail, Google Calendar, DocuSign, FactSet, WordPress, LegalZoom, MSCI, SimilarWeb, Harvey, Apollo, Clay…

‍

🧩 Specialised plugins by department: finance, HR, legal, design, engineering, operations, investment banking, equity research, private equity, wealth management

‍

🏪 Private plugin marketplace: administrators can create an internal catalogue and control which plugins each team can use

‍

🔄 Multi-application orchestration: Claude can chain tasks between Excel and PowerPoint, carrying the context from one application to the other

‍

Book your 30-minute call now

A no-obligation conversation to build your AI training around Claude.

👉 Tell us about your AI project 👈

‍

New Claude feature no. 4: Claude Code Security

‍

Release date: 20 February 2026

Anthropic is launching a tool dedicated to cybersecurity. Claude Code Security scans codebases to identify vulnerabilities and suggest targeted fixes.

For tech teams, it is a considerable time saver on security audits.

‍

Claude model comparison — February 2026

Criterion	Claude Opus 4.6	Claude Sonnet 4.6	Claude Haiku 4.5
Release date	5 February 2026	17 February 2026	October 2025
Price (input / output per M tokens)	$5 / $25	$3 / $15 best value	The most economical
Context window	1M tokens beta	1M tokens beta	Standard
Computer use (OSWorld)	Very advanced SOTA	72.5%	No
Extended thinking	Yes	Yes	No
Autonomous agents	Yes — up to 14h+ of autonomy record	Yes — multi-step tasks	Limited
Office integration	Excel + PowerPoint + Chrome	Excel + PowerPoint + Chrome	Via API only
Cowork (enterprise plugins)	Yes (Team & Enterprise)	Yes (Team & Enterprise)	Limited access
Available connectors	Google Workspace, DocuSign, FactSet, Harvey, MSCI, WordPress…	Google Workspace, DocuSign, FactSet, Harvey, MSCI, WordPress…	Via API
Coding & security	Claude Code + Code Security new	Claude Code + Code Security new	Basic code
Main target	Complex tasks, long-running agents, audits, refactoring	Versatile: analysis, summaries, strategy, everyday use default	Real time, volume, chatbots, quick tasks
Availability	claude.ai + API + AWS / GCP / Azure	claude.ai + API + AWS / GCP / Azure default model	claude.ai + API + Cloud

Key takeaway: Sonnet 4.6 offers the best performance/price ratio and becomes the default model. Opus 4.6 remains the choice for the most complex tasks and long-running autonomous agents. Haiku 4.5 is optimised for volume and speed at lower cost.

‍

Key takeaway: Sonnet 4.6 offers an exceptional performance/price ratio. Opus 4.6 remains the choice for the most complex tasks. Haiku 4.5 is built for volume and speed.

‍

What each new Claude feature changes for you day to day

‍

For a finance team: load a complete report, ask for an analysis of variances against budget, and get a summary ready for the executive committee.
With Cowork, Claude connects to FactSet and generates the PowerPoint slides directly!

For a legal team: load a contract + its amendments + the case law, and ask "what are the risks in the non-compete clause?"
The Harvey and LegalZoom connectors strengthen the analysis.

‍

For a marketing team: brief Claude on the entire annual strategy, then ask it to adapt messages by channel and by target while keeping overall consistency.

‍

For an HR team: use the Cowork HR plugin to write job descriptions, screen CVs, organise recruitment processes and analyse annual reviews.

‍

For a tech team: Claude Code + Claude Code Security to code, audit and secure codebases.
Cowork takes over for non-technical tasks.
‍

The line between "conversational assistant" and "operational agent" is disappearing.

And that is exactly what we see in training: teams that master these tools reach a new level of productivity 💡

‍

Book your 30-minute call now

A no-obligation conversation to build your AI training around Claude.

👉 Tell us about your AI project 👈

‍

Why is everyone talking about Claude right now?

‍

The impact is such that financial markets are reacting. Shares in many SaaS vendors fell following Anthropic's announcements, and some investors fear that AI agents will make traditional products obsolete.

Anthropic's valuation rose to $380 billion after a $30 billion funding round.
The company already generates 80% of its revenue from the enterprise segment.

As Kate Jensen, Anthropic's head of the Americas, puts it: in 2025, Claude transformed the way developers work.
In 2026, it will do the same for all knowledge work!

‍

Book your 30-minute call now

A no-obligation conversation to build your AI training around Claude.

👉 Tell us about your AI project 👈

FAQ

Answers to your questions about AI in business.
How to train your teams, identify the right use cases and roll out AI effectively.

Book a call

How can I train my teams in AI without it taking months?

It is the question we hear most often. The short answer: with the right method, a single day is enough for your teams to really start changing their habits. At AI Sisters, we have built a three-step approach, tested on more than 350 projects:

Step 1 - Understand, not just "discover"
We start with an AI Show: a live, immersive session with real-time demos on real cases. No generic slides. Your teams leave with a clear view of what AI can (and cannot) do for them.

Step 2 - Practise on real topics
Hands-on workshops are organised by role: HR, Finance, Marketing, Sales, Legal, Ops. We use your real documents, your real emails, your real tools. Prompts are tested live.

Step 3 - Embed new habits at D+30
One month later, we run an impact review: what has been adopted, what is still blocking, what needs adjusting. This follow-up is what makes the difference.

In short: training your teams in AI does not take months. With an approach built on immersion, role-based practice and impact follow-up,

Can AI training really be adapted to my sector (healthcare, finance, industry...)?

Yes, and that is exactly what separates useful training from training people forget.

Before each project, AI Sisters carries out a pre-audit: we analyse your typical documents (contracts, emails, reporting, product sheets), your tools (Office suite, Google, CRM, ERP) and your workflows.
We identify where AI can create value in your specific context. Then we choose, from our 40 business experts, the one who knows your sector best.
An HR team in an industrial group will not see the same examples as a B2C sales team. Prompts, exercises, use cases - everything is adapted.

‍In short: AI Sisters adapts each training programme to the reality of your sector, thanks to a pre-audit and 40 business experts. Healthcare, finance, industry, services: content is built on your real documents and your real constraints, not on generic examples.

Can AI training be funded by my OPCO?

Yes, in most cases.
AI Sisters training is Qualiopi-certified, which makes it eligible for OPCO funding (French training funds). The level of funding depends on your skills operator, the scheme used and your available budget, but in many cases a significant part of the cost can be covered.

In practice, we support you through the whole application: detailed quote, compliant programme, supporting documents, follow-up until reimbursement.

‍In short: thanks to Qualiopi certification, AI Sisters training is OPCO-fundable. The amount varies with your operator and your situation, but it can mean substantial savings.

Is AI training better on-site or remote?

Both work. What matters is the intention behind the format.
‍
On-site is ideal for a strong kick-off: a seminar day, a moment of collective co-creation, a team starting out together.

Remote is perfect for multi-site or international teams - our online sessions use training assistants via chat to keep things interactive (Q&A, co-writing prompts, live exercises).

Either way, the method stays the same: AI upskilling + practice + follow-up. And every session is led by a business expert, not a generalist trainer.

‍In short: on-site or remote, the format adapts to your organisation. AI Sisters keeps the same level of quality and interactivity in both cases, with business experts dedicated to each session.

Which tasks can I really hand over to AI in my team today?

The best way to answer is by role.

‍HR: screen CVs, generate job descriptions, write emails to candidates, prepare interviews.
‍Finance: explain an Excel table, write a commentary on results, prepare a forecast budget.
‍Marketing / Comms: create an editorial calendar, adapt a message into several formats, A/B test hooks.
‍Sales: write prospecting emails, personalise sequences, prepare responses to tenders.
‍Legal: analyse an NDA, flag risky clauses, generate a first draft to review.
‍Leadership: summarise committee meetings, prepare strategy notes, build impact scenarios.

In our AI workshops, we do not show fictional cases. We take your real documents and test AI on them, live.

Before AI: 2 hours to write meeting minutes.
With AI: 20 minutes.
‍
Before AI: 3 hours to analyse an Excel file.
With AI: 30 minutes.

‍In short: AI can speed up dozens of practical tasks in every department. AI Sisters helps you identify the ones with the most impact for your team, and test them straight away on your own documents.

My team is not technical. Can they still use AI?

Yes. In fact, that is the whole point.

Our approach is 100% role-based, with no technical jargon. We start from your teams' real problems (lost time, repetitive tasks, heavy reporting) and show how to turn them into simple prompts.

At the end of the workshop, each participant leaves with ready-to-use prompt templates adapted to their job.

The goal is not to train data scientists. It is for every employee to be able to say: "This is how AI helps me in a typical day."

‍In short: You do not need technical skills to use AI every day. AI Sisters turns the tools into accessible, role-based practices, and gives every participant prompts they can use the very next day.

How can I be sure AI does not make mistakes in a sensitive context (legal, medical, financial)?

It is a more than legitimate question, and we address it directly in our training.
We do not just show what AI can do. We also stress what should not be delegated to it, how to review and validate its answers, and how to combine a clear business brief with a well-built prompt to get the best quality.

Our experts know the constraints of your sector: compliance, deadlines, accuracy, GDPR, approved tools.
Nothing is left to improvisation.

‍In short: AI Sisters builds the limits and risks of AI into its training. Your teams learn to use AI reliably, knowing what they can hand over to it and how to check the results.