My team will continue to use AI to support us in creating copy briefs, and in updating existing copy, but we won't be using it to write for our clients. Our human writers are much better.

I expect that the number of people considering using and starting to use AI tools will increase as more and more use cases are discovered and advertised for these types of tools. There are multiple ways such tools can work in tandem with writers and editors.

However, without active participation and understanding of these technologies, I believe the barrier to entry is still quite high for many people, as well as the novelty of the technology for the purposes of content writing is a contributing factor, putting people off.

This validates my comment above – marketers are open to the idea of AI, but they need to see more about its applications.

Content marketers are rightly feeling defensive as AI is often marketed as their creative replacement. Still, they're open to it. And they'll be even more open to it once they can see its applications beyond just a replacement to their writing ability.

I think this is interesting when you consider how many people don't currently use AI tools because of their quality. Clearly, most people think they're going to improve a lot in the coming years and become more ubiquitous—or they're just not sure and selecting 'Maybe', because, well... maybe?

This illustrates a possible disconnect. Most content marketers view AI as only a tool for producing content. But, from what I've seen as recently as the chatGPT craze, there are applications around performing research, outlining your content, identifying keywords to target, etc., that can be handled entirely through AI.

Imagine freeing up time from all of that more administrative work to focus more on the creative?

These things will help to improve content quality, not lower it.

I believe the responses to this question provide a good direction for the needs of users, which can be incorporated into the research and development of new technologies. Considering the infancy of the niche and the fast development of models in this space, it is difficult to foresee the timeline of when AI content writing will reach some of these demands, however, I do not think any of them are impossible standards to meet.

This higher quality output could be just around the corner. Almost all of the tools mentioned by respondents use the GPT3 algorithm. GPT3 is impressive, but already dated. GPT4 is expected to be released in the next few months and likely to be nearly 600 times as powerful (measured in terms of parameter count).

Google's LaMDA is already available in limited form and producing output that appears significantly more "human" than we are seeing from the commonly used tools. Those waiting for higher quality output to jump into AI might not have to wait long.

The lack of quality in AI writing comes down to significant misunderstandings: it doesn't understand the value of truth, it doesn't understand context in many topics, and it doesn't understand a target audience. While these problems still persist I couldn't be encouraged to use AI.

In the future, AI will be part of our day-to-day. In the same way that spell check has become a part of our day-to-day, AI is going to do the same.

When you want an image for a blog post, you will rely on AI because it's easier to use and you don't have to worry about copyright infringement, or any of those other legal challenges that might come from purchasing or finding a stock photo.

In addition, AI will be used to simply create better stories. Yes, humans will always be a part of the process, but if an AI can support you in developing a story faster and more efficiently similar to spell check, of course, we're going to embrace it.

I believe when you fast forward 5 years from now, we are going to see that 90% of marketers and creators are using AI in some format to deliver content that is excellent.

We have moved past 'Is this gonna come or is this not gonna come?' And we've moved to, 'How disruptive is this going to be?' The speed at which this is changing is not the speed at which, say, browsers were getting released in 1995. I can't say, 'Well, Chrome was 20 years away from Netscape Navigator, so we've got 20 years.'

This might be 20 months. We have more computing power, we have more people, and things are going faster, so don't expect it to move at a snail's pace.

While I wouldn't be happy to use AI content at the moment, it has definitely improved in recent years so I'm sure some organisations will use it. Whether that's a good idea is another thing.

This data is positive in terms of managing expectations regarding the process of content production. I believe that it's also important for agencies and content creators to be transparent on how and why they incorporate such tools in their processes to help build trust with their clients, whilst at the same time being honest on the need for such technologies and their role.

People could start adopting AI tools today, and whether they do or not is more about a philosophical stance than the actual capabilities of the technology. Do you want AI-generated content? If you do, the tools are out there already, if you are savvy and aggressive enough to go use them.

Interesting data, as it demonstrates the reality of how difficult it is to foresee the future of this sector.

I think an equally valid and important question to ask is: What is the role of AI-generated content and human-generated content? This can help put into more context the tandem in which AI tools and humans can work together for content strategy implementation.

An equally valid question to ask is also how can such tools be used in content strategy planning, as there is inherently more data in this aspect of this area of SEO.

Personally, I think 3-5 years is a touch on the conservative side. The pace of development in this space is incredible, and I'd be very surprised if it took another 3 years for us to get to the point where the majority of digital marketers are—at the least—supplementing their content creation with AI tools.

I think AI content will be used increasingly in content marketing, but I feel (and hope) that we will always need humans involved in the process to ensure that writing is high quality, engaging and written with the aim of helping people. Humans have complex needs, and I think humans will always be needed to make sure content addresses them.

I think we’re already there. There are some AI writing tools putting out perfectly fine, serviceable content (at least after fact-checking and editing). And there are quizzes showing most people can’t tell the difference.

But you know what AI can’t do? Draw great ideas out of someone through conversation, distil those ideas into their essence, put a unique spin on that essence, and turn it into excellent copy.

I also think that AI writing will never quite replace human writing, especially in certain styles or niches. (If I could train a Punchline Copy AI to generate funny conversion copy for my first drafts, would I? No, I would just give it a name and ask it to keep me company.)

This might be a low bar. Most of the content found on page one of search results these days is pretty robotic and formulaic. In fact, I bet most of the content generated through AI would actually be an improvement over this content.

For everyone else, creating content with a clear and differentiated POV, original insights from others, original/proprietary data, etc., they'll never have to worry about being indistinguishable from everyone else.

If anything, this should be a call-to-action for brands everywhere to create to be different, not to (continue) to create to rank, drive traffic, etc. Because yes, AI can create that type of content, too.

For some content, we're already here. The interesting question is whether there are any types of content that AI will never be able to produce indistinguishably from a human, and what those might be?

In really good long-form copy I'm not sure this will ever be possible. In shorter, generic stuff without a strong tone of voice I think it could be possible within five years.

In my experience, finely tuned A.I. is already as good as humans, especially given specific constraints, e.g. "Write a 150-character tweet about why cash accounting is better than accrual accounting."

There's a solid majority here who think AI content will become indistinguishable, but almost 1 in 3 people thinking the opposite was higher than I expected. AI content tools are still new technology and they're getting close—it doesn't seem realistic to think it will never get there.

The longer the content you want, the farther out you need to be looking. The shorter form, more factual pieces can be written pretty much indistinguishably from humans today. If you're talking about opinion pieces or real, in-depth reviews, there's always gonna need to be a human touch.

In really good long-form copy I'm not sure this will ever be possible. In shorter, generic stuff without a strong tone of voice I think it could be possible within five years.