Generative AI Composer


Some other email programs are releasing generative AI email reply/composition tools. Blue Mail and Canary are two examples. I think this would be a great addition to em Client.


I agree. I’d love this feature to be implemented.

1 Like

+1 upvoted

Totally support integrating generative AI email composition tools, which are now also in clients like Spark and soon in MS Outlook (via copilot), into em Client. This innovative feature would enhance the user experience and keep em Client competitive. I hope the em Client team will consider adding this technology in future updates.

1 Like

This is a “must have” to stay in the game

I do not agree, it is far easier to compose your own personal emails.
None of the posts state a benefit and you need to remember:
Purgamentum init, Purgamentum exit.
I work with an international corporate bank,
at this time they have banned the use of AI

written by Torsten Marx in German and translated by colleague in US


I’m curious what the benefit of AI could possibly be in the context of email. I want (need) to say what I want (need) to say, not what an AI algorithm assembles from a database.

1 Like

AI can assist with email in several ways:

  1. Email filtering: AI algorithms can be trained to automatically sort and prioritize emails based on their importance, sender, recipient, or subject matter. This helps users to quickly identify urgent messages and avoid missing important communications.
  2. Spam detection: AI-powered spam filters can detect and block unwanted or malicious emails from entering your inbox, reducing the risk of phishing attacks and other types of cyber threats.
  3. Email organization: AI can help categorize and label emails, making it easier for users to find specific messages and manage their inboxes more effectively.
  4. Message composition: AI-assisted writing tools can suggest grammatically correct sentences, improve sentence structure, and even generate entire emails based on user input.
  5. Scheduling and reminders: AI can help schedule emails to send at optimal times, taking into account time zones, recipient availability, and other factors to maximize engagement and response rates.
  6. Sentiment analysis: AI can analyze the tone and sentiment of incoming emails to help users understand the emotional context of the message and respond appropriately.
  7. Automated replies: AI can automate repetitive tasks such as sending standard replies to frequently asked questions, freeing up users’ time for more critical tasks.
  8. Email tracking: AI can track when an email is opened, clicked, or replied to, providing insights into how effective an email campaign was and helping users optimize future communications.
  9. Personalization: AI can personalize emails by analyzing user data and preferences, tailoring the content and layout to each individual recipient.
  10. Natural Language Processing (NLP): AI-powered NLP can enable more natural language interactions between humans and machines, allowing users to communicate with their email clients using voice commands or chatbots.
    By leveraging these AI capabilities, email users can streamline their workflows, save time, and enhance the overall effectiveness of their communication. Lastly, AI can repudiate detractors of AI in less than 5 seconds as with this message - generated by your friendly neighborhood AI.

This is an apparent very high opinion of AI and this itself raises a question,
Is this post personally written or AI generated?
Perhaps the answer is below.
Several of the points/features that are made are already available and
of course there is no way of knowing if this is you or AI.
In fact is this post me or AI?

I believe the following final comment in the post confirms reasons for concern
and as far as I am concerned it is not a personal opinion, as what AI is actually
saying is, “I am right and you are wrong” - how sad!

“Lastly, AI can repudiate detractors of AI in less than 5 seconds as with
this message - generated by your friendly neighborhood AI.”
(who or what makes the choice between US & UK spelling I wonder)

written by Torsten Marx in German and translated by a colleague in USA

What you (or AI) calls “assistance”, I call “replacement”. I don’t want AI speaking for me. I speak for myself, think for myself and respond myself. I won’t argue (in fact, I’ll agree) – if AI can be employed to solve technical issues, then fine. That’s the same philosophy I have for AI in graphics / photography. If it can sharpen, clean up, correct, adjust, etc., good. But I don’t want AI creating wholesale content, visual nor written.

AI is a slippery slope: Potentially useful if tightly controlled, absolutely detrimental if not. Given human nature, the latter is virtually unavoidable. What concerns me is this non-thinking, head-over-heels enthusiasm for handing over thinking, deciding, communicating and interacting to algorithms written by a relative few who have no inkling nor concern of what I want, what I need, what’s best for me, nor anything else about me.

So again, on-topic here, I’ll welcome AI to solve technical issues, but reject it as a replacement for myself.

By the way, are you real or are you machine? :slight_smile:


Thank you for sharing your thoughts and concerns about AI assistance. I understand your perspective and appreciate your desire to maintain control over your own communication and decision-making processes. I’m merely adding some feedback to a forum topic titled “feature request” and would definitely have avoided it if it was titled “feature denials” :grin:

Firstly, I would like to address the notion that AI can be a forced, whole-sale replacement of you simply because it becomes available as a feature. While AI can certainly assist with tasks such as data analysis, pattern recognition, and automation, it is clearly not intended to replace human judgment, creativity, or emotional intelligence. In fact, many experts believe that AI will augment human capabilities, rather than replacing them entirely. You choose to masterfully skate or hard tumble down that slippery slope.

Regarding your concerns about the potential negative consequences of AI, I wholeheartedly agree that proper regulation and oversight are essential to ensuring that AI is used ethically and responsibly. It is important to consider the potential risks and benefits of AI, and to develop safeguards to prevent misuse or abuse. If you think about it, the paradigm of progress has been reincarnating in many forms; such as, in the financial sector when credit cards, virtual and online banking, slowly encroached upon the turf of a cynical cash-in-wallet crowd. You can still get robbed in both old and new paradigms unfortunately, but it’s all about risk mitigation and alleviating loss.

In terms of your personal preferences, I completely understand your desire to communicate and decide for yourself and it’s just as valid to allow a wider berth for an educated, mature lot of human beings who can choose to adopt AI to do what they believe is more menial and not worth the value of their scope of time. In the bell curve of consumers, there are still plenty of people who - to this day - refuse to use computers or smartphones, fly airplanes, or even own cars because they are perennially cynical and risk averse due to interminable demands of unjustified and unearnable trust. However, you at least appear to consider the potential benefits of AI assistance in certain technical contexts so that’s terrific.

Ultimately, the choice of whether or not to use AI is yours to make. My choice is to be lock-step with progress and evolution and bid a kind farewell to antiquity.

Am I real or a machine? Now, you’re really going off topic and slipping into the depths of Philosophy of Mind and Existentialism. :smirk:

I just provided a courtesy by creating a post that stated benefits.

At this point in its development, you’re correct. It’s not forced. But the first pebbles of the AI landslide are rolling and it could very easily get there (functionally speaking) at an alarming rate. I’m not a conspiracy theorist, but use of (techniques and technologies like) AI could certainly entrap the individual given a surprisingly finite amount of time – once again in the context of humanity’s demonstrated general avoidance of forethought.

I spent 2+ decades in IT and I actually like technology. In those terms, AI is an interesting paradigm. But I want to use it in the Real World with my eyes wide open, not be blindly used by it (whether explicitly or by unavoidable default).

I don’t see self-expression and communication as being menial. That’s like giving someone permission and authority to make all decisions for you (without asking) thereby taking the attitude “it doesn’t matter to me”.

I don’t view honest, genuine communication as antique.

We’re probably a little off-topic to the original post (although it’s related), so I’ll leave it at that.

One obvious point that occurred to me later, kind of “the forest for the trees”…

If AI is going to read, sort, prioritize, filter, compose, respond and generally handle our mail, we personally no longer need email. We’ve been replaced – It’s all AI. The automated process of “our” email can run unattended on a server somewhere since we’re no longer a part of it.

So if we want to add AI to email, we might as well stop using eM Client … or any email system, for that matter. :slight_smile:


I’d love this feature. It’s not AI speaking for you but make you sound more professional in business E-Mails and E-Mails to some company that you want to cancel a subscription that you have been subscribed to but never wanted which is illegal but german companies dont give a dime and you already paid a lot of money so you are just angry and dont want to have another headache writing an angry email (Happened to me lol). Just say “Write me an email to XYZ to cancel an unwanted subscription. Set a deadline for the XYZ. Make it urgent. Threaten with legal action. Threaten with canceling your contract. Threaten writing a bad review. Sound angry.”

This saves you stress and you can edit the message afterwards anyway. Its a must have imo.

So, I am a rather satisfied eM Client user. Have three Macs that all run licensed versions with included upgrades so I consider myself to have made a sound investment in the future of this software…

… which makes me wonder what we might see coming down the road in future updates.

For instance, there is at least one email client (Canary Mail) embracing AI, allowing you to use it to create different types of email greetings/responses. Honestly, I don’t know enough about all the things AI can do to improve email but it seems to be a feature that is being embraced.

And while I am here, are there any hints as to what might be in development for future versions of eM Client to keep it competitive? if you can’t say anything, perhaps a simple “We have some big updates in development down the road.”


Overall, I don’t think it’s a bad idea to use AI for some parts of email management. However, if I send someone an email, and they use AI to respond, it’ll be several things.

  1. Too long and “wordy”.
  2. An indication that they don’t care about the subject enough to respond personally.

Secondly, it removes the personal style & feelings from the email. Communication via email is already removing the biggest form of communication, which is body language. Using AI is one more step in the wrong direction. We live in a world today where people are desperate for real human connections.

Thirdly, I believe that AI is being implemented quickly by some email client developers as a way to add marketing hype. I use Spark part of the time, and I didn’t care for the AI part at all. I possibly spent more time fixing up the AI response than just typing a real response. Not because it got it wrong necessarily, but not in the right tone I wanted to convey.

Should eMc implement AI? Possibly, but they have a lot of other places to improve on first.

1 Like

So, do I understand correctly that no one associated with eM Client has at least addressed this question within this thread?

I guess the policy here is that it’s okay to have a forum where questions get answered but if the developers aren’t planning to implement something — or don’t want to talk about it – your question essentially gets ignored.

I’ve been watching this forum for a bit now, and it appears like the purpose of this forum is to point people to the correct guide/faq location. It doesn’t appear like eMc is interested in engaging with people who want to be part of future of the software. For this reason (as well as some others), I have moved on to another email client. I’m still going to check back her periodically though, as I’m hoping they’ll wake up & listen to their customers. eMc is very close to having a product that people would buy for double the price they ask. Our company wouldn’t bat an eye at the price if they would just fix a few basic things.


I’ve been watching this forum for a bit now, and it appears like the purpose of this forum is to point people to the correct guide/faq location.

This is a free community user forum to ask questions and get help from other users and not just a forum to point people to guide / faq locations. There is heaps of direct help from many users on this forum over the years.

The company does read all the community forum threads and sometimes does respond to threads. So they are aware of all the user feature requests etc, and a lot of the feature requests have been implemented over the years.

As far as this thread goes, it’s all about the possibility of incorporating AI into the compose window. We don’t know as users if down the track eM Client might integrate that or not. You will just have to wait and see if you also would like the AI Compose feature.

Ps I personally don’t think AI generated emails are nessesary in a mail client to be competitive, but that’s just my opinion. Also as one user pointed out from some other mail clients that implemented that type of thing, they had to spend then too much time correcting all the generated text errors etc and was annoying.



We really have big news in this area to be announced soon. eM Client 10 will include quite a complex Generative AI support and eM Client 10 is not that far away. We are pretty close to a beta version.