Evernote still hasn't done anything with AI like Notion has done.

[Anthony2023 1]

Evernote still hasn't done anything with AI like Notion has done. While Evernote has some basic AI features such as text recognition and OCR, they are far less comprehensive than the AI capabilities offered by Notion. Notion's AI-powered features include automated text analysis, document organization, search capabilities, and integrations with other applications. Additionally, Notion AI also offers support for private conversations, allowing users to have secure conversations without worrying about their data being seen by others.

[PinkElephant 8,158]

Uhhhh - now everybody needs AI to be cutting edge, sexy, forward looking or at least  in the game.

Ever tried the automatic filing of emails forwarded ? Or of web clips ? As with every AI, you need to invest some initial training effort. This because EN won’t touch your data themselves, your privacy has priority.

[agsteele 2,952]

@Anthony2023 Sounds like you know which application meets your needs best.

[gazumped 11,693]

...and wasn't that launched a coupl of weeks ago "after several weeks of testing"?  Evernote are currently in the middle of a takeover by an AI-specialist company,  so when the ink is dry on the deal they'll presumbly inherit some expertise..

[Boot17 1,465]

This post brought to you by the ChatGPT conglomerate. You should be using AI everywhere... (FWIW - I had an AI powered meal for breakfast. /s)

I have found ChatGPT to be extremely helpful, but I don't want gimmicky AI implementations in Evernote just like I don't want Evernote to implement the kitchen sink feature-set. Quite often things are done better in a separate / dedicated program. There is probably going to be useful AI related features in Evernote eventually (like @gazumped mentioned because of the new owners), but this is still a brand new game. I wouldn't be surprised if we see some Bending Spoons initiatives on implementing some kind of AI-like things in the coming months since AI is so hot right now and they are already in that space. They could see it as a big win and a reason to justify charging more $$€€.

[Dave-in-Decatur 3,970]

@Anthony2023, I apologize in advance for my suspicious mind. But you sound like an AI-generated post from Notion's marketing dept. Hoping you're not (I suppose some people sometimes think my posts are evidence of Evernote's poor AI tools), but hard to see why someone would start a new thread with this in an Evernote support forum.

[eric99 984]

On 3/6/2023 at 8:40 AM, Anthony2023 said:

Evernote still hasn't done anything with AI like Notion has done. While Evernote has some basic AI features such as text recognition and OCR, they are far less comprehensive than the AI capabilities offered by Notion. Notion's AI-powered features include automated text analysis, document organization, search capabilities, and integrations with other applications. Additionally, Notion AI also offers support for private conversations, allowing users to have secure conversations without worrying about their data being seen by others.

Evernote intended to do it about 9 years ago and the whole evernote community strongly objected for privacy reasons. So EN buried their plans immediately...

[gazumped 11,693]

2 hours ago, eric99 said:

evernote community strongly objected for privacy reasons

Actually I think it was more a throwaway "selected staff may view your notes for admin reasons" that threw everyone,  not AI itself - but you raise a fair point.  ChatGPT (on which I think Notion's AI is based) shows this in FAQs...

Quote

As part of our commitment to safe and responsible AI, we review conversations to improve our systems and to ensure the content complies with our policies and safety requirements.

ChatGPT General FAQ | OpenAI Help Center (Q5)

Maybe we should be careful what we're wishing for...

[PinkElephant 8,158]

The unanswered question is whether people are wishing here to have AI help with content or with the way the user is organizing it. Plus whether it should be a server based function (the sort of "other users with similar content do this or that differently") or a device based function which would be superior in privacy.

To tell "we need AI because everybody has it now" is an invitation to get a marketing claim, nothing else. AI is not "intelligent" just because intelligence makes up 50% of the buzz word.

 

[gazumped 11,693]

7 hours ago, PinkElephant said:

AI is not "intelligent" just because it makes up 50% of the buzz word.

Totally agree - just read someone else's definition of 'smart' filing which includes their AI putting notes in categories based on content.  Anything containing pictures or words relating to wooden furniture forinstance would automatically go into a category.  So your picture of grandad in his favorite chair would go into 'wood furniture' along with the rest. That sort of instance could give rise to as many correction actions as is saved from just manually titling and tagging the notes in the first place...

I have a deep suspicion of 'black box' curation. 

[eric99 984]

13 hours ago, gazumped said:

Actually I think it was more a throwaway "selected staff may view your notes for admin reasons" that threw everyone,  not AI itself - but you raise a fair point.  ChatGPT (on which I think Notion's AI is based) shows this in FAQs...

ChatGPT General FAQ | OpenAI Help Center (Q5)

Maybe we should be careful what we're wishing for...

At the other hand, this could have changed Evernote's future 9 years ago. It was a revolutionary idea for a note taking app at that time...

[Dave-in-Decatur 3,970]

21 hours ago, PinkElephant said:

The unanswered question is whether people are wishing here to have AI help with content or with the way the user is organizing it.

I can see why some minor use of AI to put notes into notebooks, or add tags, could be helpful, if your system was fairly simple and the contents of new notes quite predictable. Evernote Web clipper already offers this with "smart filing" of new clips, which I find offers the right notebook maybe 2/3 or even 3/4 of the time. But it's never something I would rely on with checking before clipping.

As for AI help with content ... I can't begin to imagine that, but that relates to my way of using Evernote and what I use it for. For someone who uses it mainly as a high-tech filing system, that might work.

[eric99 984]

The major question is, do you really want your private note data to be integrated into the centralized bot's self learning neural network? I definitely don't want so. 

Before you know it, queries will (indirectly) leak  note data to each other  because AI bots don't always know their sources anymore  😏

 

[gazumped 11,693]

1 minute ago, eric99 said:

do you really want your private note data to be integrated into the centralized bot's self learning neural network?

A lot of users were upset about a small number of actual humans having access to their data,  but I'm not sure how widespread anti-AI feeling is going to be - where's the difference in your data being stored in digital form -vs- because of a particular pattern in that digital data it gets flagged as misspelt or needing to be moved to another notebook...?

My favourite science expert Sabine Hossenfelder had a think recently about whether AI chatbots actually understand what they're processing - despite the blurb her conclusion seems to be that AI has a limited / partial understanding.

(My attitude to privacy is a little complicated - I started tech support back in the days when we could see users' accounts - passwords and all - in our admin screens,  and one gentleman asked me to read a couple of his emails to him since his computer was broken.  While that sounds a little wild,  it wasn't totally insecure - one of my colleagues got fired on the spot for writing down frequent-caller user details so he could "save time"...)

 

[Dave-in-Decatur 3,970]

Define "understand". As opposed to "misunderstand". Or "I think I understand". State number of consecutive years in a monogamous relationship to confirm knowledge of the difference.

[PinkElephant 8,158]

@eric99 posted an interesting observation. AI is usually trained using a huge volume of data. It derives metainformation from the training data it is fed, and it does so in ways that can’t be understood or reproduced. This means that a technically identical AI hardware, fed with the same set of data will build a different representation of the data. This is one of the big issues with AI (and explains why genius and idiocy reigns side by side), the inability to QA the outcome.

You can run it a million times, and you get a million different results. Often the „best“ results (in terms of stability) are as well the ones that carry the most baggage in form of „prejudice“ - AI bots have been proven over and over to be biased, worse than any living racist, homophobe or psychopath.

To train an AI to assist EN users with say filing, tagging, creating good note titles, write better searches etc. one must feed the AI with a huge volume of data - like the data of ALL users. This has 2 implications: The bad practice is probably by far overwhelming, the good practices are related to use cases that are implicit and not declared and even in the best of all worlds part of the data will burn itself into the neural network, starting a second life there. This is not in the meaning that it is computer readable like in a database, but it is integrated into the AIs brain, building the patterns stored there.

My conclusion: If EN would go down this trail, I would like to know exactly what they do, how they do it and if the implications were thoroughly analyzed. Or a little checkbox in account settings defining my account as a no-go zone for any AI bot.

[gazumped 11,693]

With the release of GPT4 I'm firmly in the torches and pitchforks group.  Why in the world would you voluntarily connect your personal information (which includes anything you say or type into it) with a global entity that is massively faster and smarter than you and any interface device (other than another similar entity) that you might be using?  We're talking superhacker smarts here and I hope browser manufacturers are ready for it.

The possibilities are genuinely frightening - especially when you consider the said device could be given deliberately misleading input information or just flat out asked to influence opinions in one direction or another.  And what about those messages from family members that sound exactly like they should / audio clips of celebrities endorsing views that you support?  Any chance a virtual 'you' might generously give your savings to an (un)deserving cause?

From this point on you can accept nothing as being factual.  Check information from multiple sources / do things in person if you can - while it's not the Singularity we're expecting,  I think our society just got disrupted big time.

[eric99 984]

6 hours ago, gazumped said:

Why in the world would you voluntarily connect your personal information (which includes anything you say or type into it) with a global entity that is massively faster and smarter than you and any interface device (other than another similar entity) that you might be using? 

That's what notion users apparently  already do 😏

[Dave-in-Decatur 3,970]

Please note that we continue to discuss a topic introduced by a "member" who still has only that 1 post and is almost certainly a sock puppet that posted content generated by Notion's AI. (At any rate @Anthony2023 has not yet chosen to prove otherwise.) Also note that none of us was fooled by it.

[Anthony2023 1]

I cannot believe that there are so many discussions.

It's okay if you think that I am a bot, but I have been an EN user for more than 20 years (This is an error, I am only a user starting at 2011 Thanks @PinkElephant) and I have paid for it for about 5 years. I am someone who loves using it and I just want EN to consider adding AI functions, as it is a trend. Please do not talk about privacy, as there is no privacy in this world. You are just fooling yourself if that is the reason why you do not like AI.

[PinkElephant 8,158]

Exaggerating like in your other posts - EN launched 15 years ago ...

If you have no privacy, you could work to achieve it. Which requires a quantum of personal intelligence. If you feel overwhelmed by the challenge, ask your AI.

[Anthony2023 1]

Thank you @PinkElephant for pointing out my error. Does this prove that I am not a bot to some extent? I am not trying to start an argument here, and I apologize if my previous message caused any hard feelings to anyone. I am a user who has moved to Notion recently, and I believe that AI is a powerful tool.  If EN thinks that this feature is useless, that's fine. I totally respect their decision.

[PinkElephant 8,158]

We agree that AI can be a powerful tool. But this can be seen from different views - the current hype around Chat GPT is ignoring some of them. Examples are false answers, answers to which the AI stubbornly sticks even if shown proof of being wrong. Another example is the lack of links to the sources, which makes it impossible to check the validity. A third issue is the typical bias of AIs - a bias that is rooted in the training material it receives to build its network, so it can't be avoided easily.

I think the current addition of the copilot function to O365 illustrates pretty well what is going wrong: You throw a little information into PowerPoint, and ask the Copilot to convert it into a full presentation. And then you need (say) 10 people to watch it, because what else it the purpose of a presentation. So first hand you save on your own effort, to then take away the life time of 10 others for yet another bloodless, word rich, meaningless hour of slides.

Don't tell me I am wrong - I had just this experience yesterday, when a guy showed a ChatGPT output as bullet points, and even dared to read it aloud, word by word. It was this last slide, and it completely killed what positive impact may have been left from the dull presentation he made before. Maybe he wanted to show "Hey, I'm the man, I even use AI". Yes, he showed it ...

But I stray OT. If you are happy with AI in Notion, stick to it. Case closed ...

[Anthony2023 1]

You have raised a very good point here that many people ignore nowadays. AI answers are not always correct. It would be very nice if we could have an improved version of AI tools for EN, and I look forward to it. Thanks @PinkElephant

[gazumped 11,693]

Seems like it may be an expensive option too... https://www.cnbc.com/2023/03/13/chatgpt-and-generative-ai-are-booming-but-at-a-very-expensive-price.html

[Dave-in-Decatur 3,970]

16 hours ago, Anthony2023 said:

I just want EN to consider adding AI functions, as it is a trend. Please do not talk about privacy, as there is no privacy in this world. You are just fooling yourself if that is the reason why you do not like AI.

Robot or no-bot, you are making the case for Evernote not to rush to add AI. "As it is a trend" is not a reason for doing anything. NFTs are a trend (or are they not one any more?). People who say there is no privacy are people who want an excuse to steal something. Or bots trying to justify their existence. If you don't think there's any privacy, why not share your email address and bank account info, maybe your DOB? Unless you are just fooling yourself.

[Anthony2023 1]

On 3/22/2023 at 4:52 AM, Dave-in-Decatur said:

Robot or no-bot, you are making the case for Evernote not to rush to add AI. "As it is a trend" is not a reason for doing anything. NFTs are a trend (or are they not one any more?). People who say there is no privacy are people who want an excuse to steal something. Or bots trying to justify their existence. If you don't think there's any privacy, why not share your email address and bank account info, maybe your DOB? Unless you are just fooling yourself.

I think you may have a misunderstanding about privacy. Can you provide an example of how AI can weaken our privacy? Has there been a case where AI has caused a privacy issue? While it is true that new technologies have often faced criticism when they were first introduced, such as the calculator, it is important to carefully evaluate the potential benefits and risks of implementing new technology, including NFTs. I agree that it is crucial to protect privacy in these evaluations and ensure that new technologies are implemented in a responsible and ethical manner.

[gazumped 11,693]

4 hours ago, Anthony2023 said:

Can you provide an example of how AI can weaken our privacy?

https://help.openai.com/en/articles/6783457-chatgpt-general-faq

5:  Who can view my conversations?
As part of our commitment to safe and responsible AI, we review conversations to improve our systems and to ensure the content complies with our policies and safety requirements.

AFAIK no live third parties are currently analysing or reading my notes...

[eric99 984]

5 hours ago, Anthony2023 said:

Can you provide an example of how AI can weaken our privacy? Has there been a case where AI has caused a privacy issue?

https://www.cnet.com/tech/services-and-software/chatgpt-bug-exposed-peoples-conversation-history/

"...OpenAI says it's unable to delete specific prompts from a user's history, adding that people shouldn't "share any sensitive information in your conversations..."

[gazumped 11,693]

Interestingly I tried to get ChatGPT3 & 4 plus the new Bard to comment on security issues when dealing with AI.  All three of them took an unusually long time to think about it and had errors when producing a response.  Bard got the furthest - it's concerns included:

1. Data Protection: Ensuring that any data used to train an AI model is properly secured and protected against unauthorized access.

2. Bias: Inappropriate bias in an AI model could lead to unfair outcomes. This can be avoided by regularly testing models for accuracy and fairness.

3. Human-Machine Interaction: AI systems must be designed in such a way that humans can safely and productively interact with them.

4. Privacy: AI systems can collect vast amounts of personal data from users and must be built with appropriate security and privacy measures in place.

5. Explainability: As AI models become increasingly complex, it is important to be able to explain how and why the AI reached certain

(Had to prompt it again to continue)

conclusions. Without explainability, it’s impossible to troubleshoot the AI or make improvements.

1. Transparency: AI systems must be transparent in the way they work and use data so that users can understand their impact and trust its decisions.

2. Auditing: Regular auditing of AI systems is essential to make sure that the models are performing as expected and meeting ethical and legal requirements.

3. Accountability: AI systems must be held accountable for their decisions, and organizations must ensure that the decision-making processes are fair and appropriate.

4. Regulation: Governments and international organizations must establish regulations to ensure that AI systems are developed and used responsibly.

10

(Stopped without any further feedback)

Not saying this is in any way an indication of a conspiracy by our AI soon-to-be Overlords but I for one welcome the new age of technology...   

[eric99 984]

32 minutes ago, gazumped said:

Interestingly I tried to get ChatGPT3 & 4 plus the new Bard to comment on security issues when dealing with AI.

this is what Bing AI answers to the question: "how can AI bots threat your privacy?"can AI bots threat your privacy?

"...AI bots can threaten your privacy by using personal information in ways that can intrude on privacy interests by raising analysis of personal information to new levels 1. However, principles of trustworthy AI like transparency and explainability, fairness and non-discrimination, human oversight, robustness and security of data processing can regularly be related to specific individual rights and provisions of corresponding privacy laws2. ..."

[eric99 984]

"ChatGPT is a data privacy nightmare. If you’ve ever posted online, you ought to be concerned" :

https://gcn.com/data-analytics/2023/02/chatgpt-data-privacy-nightmare-if-youve-ever-posted-online-you-ought-be-concerned/382718/

[Anthony2023 1]

2 minutes ago, eric99 said:

"ChatGPT is a data privacy nightmare. If you’ve ever posted online, you ought to be concerned" :

https://gcn.com/data-analytics/2023/02/chatgpt-data-privacy-nightmare-if-youve-ever-posted-online-you-ought-be-concerned/382718/

@eric99 Can you give an example of what kind of data has been leaked by ChatGPT? Or are you just worried? If you are genuinely concerned, you should consider stopping the use of Google, which already sell your data for profit. 

[eric99 984]

20 minutes ago, Anthony2023 said:

@eric99 Can you give an example of what kind of data has been leaked by ChatGPT? Or are you just worried? If you are genuinely concerned, you should consider stopping the use of Google, which already sell your data for profit. 

You've probably missed my previous post: chatGPT exposed the titles of  past conversations  to other users:

https://www.cnet.com/tech/services-and-software/chatgpt-bug-exposed-peoples-conversation-history/

Also, the other article I posted  is a perfect summary of all privacy /legal problems with AI: https://gcn.com/data-analytics/2023/02/chatgpt-data-privacy-nightmare-if-youve-ever-posted-online-you-ought-be-concerned/382718/

[Boot17 1,465]

27 minutes ago, eric99 said:

You've probably missed my previous post: chatGPT exposed the titles of  past conversations  to other users:

I don't doubt that AI has already opened and is going to open up a whole big can of worms, but that particular issue wasn't really an AI-specific thing though was it? It was a bug with an open source tool that they were using and similar cross-over of account information could happen to any type of software service -- even those not using AI.

[Dave-in-Decatur 3,970]

https://www.npr.org/2023/03/22/1165448073/voice-clones-ai-scams-ftc

Someone calls you. It sounds exactly like your sister. She is in a desperate situation and needs your help. And it's someone using AI trying to rob you.

I'm sure it's just media bias, but I'm still waiting to hear of any breakthrough thing that AI accomplishes that is both ethical and legal.

[AlbertR 542]

Funny - I just saw the discussion and asked GPT itself if it created the initial post. Here's the answer:

• It's possible that a similar text was generated by me, as a language model, as I have been trained on a vast amount of text data and can generate a wide range of outputs.

 

A politically correct answer that tells us nothing - but also shows what GPT just can not. To get a more precise answer, GPT told me to give some more background information... OK, it got all (that I know) and answered:

• I can generate text that appears to be written by a human, but I cannot confirm the authorship of a specific text without additional information. The text you provided could have been written by a human expressing their opinion or generated by a language model like me to spark a discussion or attract attention to the topic.

From my point of view, AI might be used only to enhance search functionality in EN. For example to find notes by a naturally worded question (not only giving wildcarded phrases) like "Where do I have been on holidays in 2009?". It would be nice if EN-AI would understand my question, figures out what might be relevant in this context and lists these notes...

 

But: To be able to answer such questions, EN-AI has to know all of my notes, has to recognize how I describe my travel habits (maybe in comparison to other user habits to learn?). For sure nothing that can be done by a EN-specific algorithm. So EN has to ask a really big machine (like GPT). This machine is based on masses of data that has to be supplied long time before such a question might be asked. Where do this data come from? From us - the users. Do we like that all of our note will be know by GPT? I think - no.

 

Conclusion: I do not see a propper use case of (foreign) AI in EN without opening user data to its knowledge base. So please do not start any immature attempts to implement "any" AI in EN. Just keep on track with slightly improving well known features like reliability, speed and availibility for ever!

[eric99 984]

2 hours ago, Boot17 said:

I don't doubt that AI has already opened and is going to open up a whole big can of worms, but that particular issue wasn't really an AI-specific thing though was it? It was a bug with an open source tool that they were using and similar cross-over of account information could happen to any type of software service -- even those not using AI.

What this proves is that chatGPT technology is to premature to be used in EN, Notion or any other real world application where sensitive data is involved.

"...OpenAI says it's unable to delete specific prompts from a user's history, adding that people shouldn't "share any sensitive information in your conversations..."

[Dave-in-Decatur 3,970]

1 hour ago, AlbertR said:

Funny - I just saw the discussion and asked GPT itself if it created the initial post. Here's the answer:

• It's possible that a similar text was generated by me, as a language model, as I have been trained on a vast amount of text data and can generate a wide range of outputs.

 

😄 I believe I saw somewhere that chatGPT had successfully passed a bar exam. So it appears to know how to give appropriately legal-and-yet-uninformative responses when a simple, "Yes, that was me" might be true.

[gazumped 11,693]

I'll confess to using AI chatbots to help me phrase a good old-fashioned Google search better - they tend to come up with the right phrases and names for particular activities or processes.  With a combination of AI-fu and searching I very quickly built a table of names and home page addresses for one project forinstance. 

I find their ability to ramble extremely annoying though - the legalese does not convey anything useful or informative,  so I think authors will probably be safe for a while yet (but look out for GPT5!). 

Amusingly again (well I found it funny anyway...) GPT3 pointed me to two quotes - both saying similar things:

"I apologize for the length of this letter; I didn't have time to write a short one." - George Bernard Shaw

"I have made this longer than usual because I have not had time to make it shorter." - Blaise Pascal

The point being that detailed explanations are easy.  Short and informative briefings are a skill that I think requires some expertise that is not quite available yet.

 

[RJLUK 307]

3 hours ago, Dave-in-Decatur said:

😄 I believe I saw somewhere that chatGPT had successfully passed a bar exam

ok two mohitos please Chatgptbarman

[niftygeek 5]

On 3/6/2023 at 12:58 PM, Boot17 said:

This post brought to you by the ChatGPT conglomerate. You should be using AI everywhere... (FWIW - I had an AI powered meal for breakfast. /s)

I have found ChatGPT to be extremely helpful, but I don't want gimmicky AI implementations in Evernote just like I don't want Evernote to implement the kitchen sink feature-set. Quite often things are done better in a separate / dedicated program. There is probably going to be useful AI related features in Evernote eventually (like @gazumped mentioned because of the new owners), but this is still a brand new game. I wouldn't be surprised if we see some Bending Spoons initiatives on implementing some kind of AI-like things in the coming months since AI is so hot right now and they are already in that space. They could see it as a big win and a reason to justify charging more $$€€.

Ironically, I did in fact have an A.I. powered breakfast this morning based on my dietary guidelines and exercise recovery issues.

I recognize how silly such a thing sounds at the same time needing to admit I set up a personality prompt for meal planner that 'speaks' in a fake French accent to help me set up a practical meal plan (and mildly entertain me at zee same time).

And it IS silly.  But undeniably also useful to me. 🤷‍♂️

[Boot17 1,465]

1 hour ago, niftygeek said:

Ironically, I did in fact have an A.I. powered breakfast this morning based on my dietary guidelines and exercise recovery issues.

I dunno, kind of sounds more like regular programming to me wrapped in an A.I. moniker (perhaps by the devs of that software), but it does sound useful. 

[gazumped 11,693]

As I understand it,  the 'learning model' basis just means that all the manuals for all the devices and processes have been read into a chat-capable database.  Want to have a power breakfast involving Pineapple?  It can instantly give you options. 

'Smart' it may be;  but this level of "AI" involved no intelligence at all.  It isn't going to take over the world directly,  but it does mean that companies are going to have a field day adding AI to their products and armchair experts like me are going to ramble on endlessly in hopes of selling their (non)-expertise.

I find it a good sounding-board / writer's block unbunger / problem solving duck substitute... but I would not willing pay for it to be added to any app - I can get it free online anyway.

[Danni P. 22]

@Anthony2023 It's coming - Just doing it better than what Notion is doing 😉 Check one of their latest blog posts - from Federico and the Evernote team. This is the section about introducing AI in Evernote: 

Artificial Intelligence (AI)

AI is a hot topic right now—and understandably so, as it brings with it plenty of concerns as well as opportunities. Concerns not just in terms of privacy and security, but also because it can prompt a nasty tendency to innovate for the sake of innovating—adding flashy features that don’t actually help the user in any meaningful way.

Rest assured that our AI features will be opt-in only, meaning your data won’t be processed by the AI model unless you explicitly direct Evernote to do so. And if you do decide to opt in, you’re still in control of your data, and we continue to keep your data safe, secure, and private by implementing the principles of privacy by design and by default.

When handled wisely, AI also offers amazing opportunities. So the challenge we’ve been working on is to introduce AI to Evernote in a way that truly makes life easier for you. With these new features, we’re confident we’ve done just that! I’m excited to share some details about them with you now.

AI Note Cleanup

The first AI-based feature, which we’ll start rolling out in May, is AI Note Cleanup. Do you ever frantically take notes in an important meeting, then spend far too long having to tidy them up after it ends? AI Note Cleanup does the boring, time-consuming bit for you. Every user has one month of unrestricted access before the feature becomes subscriber-only.

Personal take: I’ve been using a prototype of this feature to clean up some of my notes for a few weeks now, and I honestly don’t know how I ever managed without it. I suspect many of you are going to feel the same.

AI Search

The second AI-based feature is especially technically demanding. Its working title is AI Search. With it, you can ask questions in natural language directly to your Evernote app via text and receive an instant answer based on the content of your notes.

Search is one of the features that’s most appreciated by Evernote users, so the possibility of improving it in a massive way is really exciting. But, as mentioned, there are privacy considerations at play, so we’re proceeding carefully and methodically. The team is hard at work developing it right now, and it will arrive in the latter half of the year.

 

Source: https://evernote.com/blog/evernote-pricing-upcoming-features-update/amp/