The rapid growth of article intelligence (AI) has led to justifiable concerns about copyright, ethics, and the future of human creativity. But it also enables individuals and small companies to dramatically increase their output. And given that AI is likely to be a feature of any self-employed work in the future, it’s important to understand how AI can help you scale your freelancing.
This isn’t intended to dismiss issues such as the training models for AI services using publicly available work without consent. And you may still decide that you would rather not use AI for personal or business reasons. But you should at least be aware of the options available to make an informed choice.
One aspect that may limit your use of AI is how any private company or client data may be fed into the system as part of your usage. So it’s worth checking how any confidential information may be used before putting it into an AI service.
Artificial intelligence is an extremely broad term which refers to the machines performing tasks normally requiring human abilities, including recognising images and speech, generating text and designs, decision making, coding and more.
It’s existed as an academic field since the 1950s, and has been widely applied over recent decades to power search engines, voice assistants, self-driving cars, recommendation systems, algorithmic training and investments, cyber security systems, agriculture, healthcare and more.
The recent release of a growing number of generative AI (GenAI) services such as ChatGPT, Bard, Midjourney, Stable Diffusion and DALL-E have prompted a sudden boom with an increasing number of businesses and individuals exploring ways they can utilise these tools for work.
Rather than simply sorting through data and highlighting anomalies, for example, GenAI learns patterns and structures from input data, and then creates new content with some degree of novelty and difference.
By training on a massive library of written content, these large language models (LLMs) can use deep learning techniques to then predict which letters and words should follow each other, allowing them to generate human-like text which can be used for articles, chat bots, emails and more.
Putting moral and ethical concerns to one side, there are a number of practical issues which can occur from using AI tools and services to scale your freelancing.
Artificial intelligence models currently work on complex pattern recognition, despite appearing to understand your inputs. Which means it can’t know when it’s prompted to generate a result which is socially inappropriate, out of context, or completely false. And it will present all information with the same degree of confidence, meaning completely incorrect information is presented as fact.
This is obviously a problem for anyone who accepts work created by, or in conjunction with, AI without appropriate checks. And particularly if you’re using it for anything where incorrect information can have serious repercussions.
There are also potential problems with AI showing any inherent bias contained in the training material. Or if relying on artificial intelligence means that you forget how to do those tasks if suddenly needed, in the same way that satnavs have reduced the ability of average people to use a map.
Human nature means we often give personalities to voice assistants and AI, but picturing it as autocomplete on steroids will mean you’re more likely to remember the current limitations.
Most freelancers will have experience of time and project management software, but a growing number of new and existing tools are plugging in AI to expand and improve the features they offer.
For example, Toggl and RescueTime will automatically start and stop a work timer based on what you’re doing on your computer or phone, and detect when you’re idle. They can also report on how you may have wasted time, block websites which could distract you, and offer coaching and prompts throughout the day to keep you on track.
AI tools can also help to generate better project estimates, roadmaps and planning, with ClickUp, Notion, Asana, Monday and more offering real-time alerts, brainstorming suggestions, custom reports, automated approvals and more. And machine learning patterns can highlight the factors and issues which have impacted previous projects, or when one task is reliant on existing dependencies which you may not have spotted.
Other systems, such as x.ai will work as an assistant which can communicate with your email contacts to schedule meetings.
One area where AI excels over humans is in data driven tasks, and large-scale analysis. If you’re a freelancer who tends to forget to turn on time tracking, or struggles with managing admin tasks and reporting, then utilising AI to help will allow you to keep clients and collaborators updated more effectively, ensure invoices are accurate, and give you more time to focus on your actual work.
The availability of ChatGPT, Bard, JasperAI, Rytr, Copy.ai and many other writing services has grown incredibly quickly. Mainly because many AI text services all use the same LLM data or API access to OpenAI’s GPT platforms, and then add extra features to try and differentiate themselves.
All of them tend to produce work of a similar quality, and although many focus on one particular area of content generation, they all tend to have comparable benefits and limitations.
The best usage for AI text generation is alongside a human writer or editor, rather than as a replacement. While the output has become fairly acceptable, particularly for short, easily-templated work such as hundreds of product descriptions, it’s also still typically generic, and often incorrect.
But it can certainly help you scale by offering useful templates, suggestions and drafting a starting point for articles. Almost every AI writing tool also integrates with grammar, style, and plagiarism tools to help you hone the initial output and your own edits, along with SEO tools if you need to improve your focus on keywords.
Almost all will give you a Google Docs-style editor to make changes, and will suggest alternatives for headlines, subheadings and more.
There’s an obvious benefit if you’re a freelancer who isn’t a natural or trained writer, and needs guidance in creating content for your own website and social media. Or to boost the output for any content services you offer, assuming you still ensure quality and accuracy is maintained.
Many articles suggest you can simply offer the output of AI as a copywriting or blogging service, but this carries some risks, including the potential for errors, or for clients to realise they could just implement AI content themselves for less.
But there are other useful areas for AI text to help you scale, including generating personalised emails for sales and marketing. The confidence of AI can help you produce a more effective pitch, along with freshening up CVs, covering letters, responses to briefs and marketing materials.
To get the best results, it’s important to spend time learning how to create and hone the most effective prompts. And to use AI as a collaborator or untrusted new contributor, who will require input and editing, whether that comes from you or a professional writer or editor.
It’s almost impossible to escape the headlines and hype around AI image generation and design tools. The idea that a relatively simple text prompt can produce photos, art, videos and even background music has captured the imagination of many people.
The most popular services include Midjourney, DALL-E and Stable Diffusion, but more are appearing every day. Just as with other AI, they’ve been trained on existing work to produce new designs, with a variety of techniques used to accomplish this.
If you’re looking to scale up your freelancing, it’s a quick and cost-effective way to produce a wide variety of imagery. And recent advancements in typography have made some services much better for logos or promotional work.
Using AI services can also help you to maintain a brand style and visual identity by ensuring rules are maintained.
As with all AI-generated content, there’s a limitation in the accuracy of the image created, particularly noticeable in human faces or limbs. If you’re asking for a specific product to feature, this may not be the one AI chooses to replicate.
The system of prompts for AI image and video generation can also be more complicated and need more refining to achieve the exact style you’re aiming for. This is one area where existing creative freelancers can have an advantage in using AI image generators, as a knowledge of photography, design and art will enable more accurate instructions.
One big area of concern is that the UK’s position on copyright ownership is ambiguous. Some countries have stated that copyright can only exist and be owned by human beings, such as America and Germany. Until that’s clarified, anything you generate through AI may not be eligible for copyright protection.
It’s also important to check the limitations on the level of licence you’re using to access an AI service. For example, if you use Midjourney through a free membership, you grant the company ownership of your work, and you’re permitted to use it under a licence, whereas you have ownership with a paid membership.
The good news if you’re already using various design tools is that they’re increasingly integrating AI functionality, including Canva and Adobe Photoshop. But again, it’s worth checking what implications this has for ownership and licensing.
Where AI can definitely save time in design is enabling non-artistic people to produce drafts and samples which can be provided to professional designers to implement. It can be hard to convey exactly what you want for a logo, website, or an advert, so being able to use AI to create a mock-up can help get your requirements across more effectively.
And it can also be used to enforce house styles if you’re working with one or more designers, ensuring that everything matches under your brand.
It’s relatively quick and easy to use AI to create content for social media, but it can also help you track, report, and manage the success of your work. Or to assist you with other areas, such as finding the right people to target with your messaging.
Tools such as Socialbakers offer AI-powered Persona Mapping to identify your target audience, and to then find influencers whose audience would match appropriately. Sprout Social has also implemented AI to enhance areas including their listening and intelligence function to help identify actionable data. Other alternatives also have similar tools available, including Audiense.
Other services such as Buffer Feedhive and Ocoya are integrating AI into their own content creation functionality to generate the text, images, hashtags, or even emojis you can choose to use or edit before publishing.
Given the fact that social media management can often be time-consuming if you’re repurposing content for different channels, using AI to suggest remixed posts which are optimised for each platform could enable you to spend more time on areas where your human input is more valuable.
And while you should question and check any suggestions for personas, target audiences and suitable influences, manually collating the required data can be immensely time-consuming, particularly if you’re working with larger brands. Even if the AI analysis simply gives you a suggestion to work from, it can be a big help.
But you need to be careful about blindly publishing AI-generated content. Not only can it be factually incorrect, but AI has no awareness around context, or sensitivity when it comes to subjects which could cause offence. As we’ve seen from automated scheduled posting, you can damage a client brand quickly by publishing at the wrong time. Especially if you’re handing everything over to AI and not spotting negative comments and responses in a timely fashion.
Increasingly AI is also being integrated into social media platforms themselves, such as LinkedIn testing AI generated suggestions for posts, or Meta testing AI-powered advertising tools to create image backgrounds and text variations at scale.
One issue in freelancing for customer-facing brands is that you can often spend time trying to resolve relatively simple customer enquiries and complaints. Chatbots are a well-publicised potential use of AI, along with automated email personalisation. And while neither is suitable for complex problems, they can certainly help with simple questions or to filter contact appropriately.
There are a variety of Chatbot platforms which can be implemented on behalf of your business or clients, providing quick responses with predefined answers to common queries, and the suitable routes for issues to be escalated to you, or your client, if necessary. And because they can run 24 hours a day, you’re less likely to wake up to an impatient customer complaining that you didn’t respond at 2am.
Many tools such as EBI.AI, ProProfs Chat, Chatfuel or Aivo also integrate other marketing and lead generation features such as broadcasting messages to multiple users, list building and drip campaigns triggered when someone completes a specific action.
This can be combined with automated email personalisation, using AI to amend defined copy and details appropriately for each user. And these can be tailored based on customer segments, the action they’ve taken so far, or any other data you’ve acquired.
And usually, these services are offered alongside content optimisation, including email subject lines, and any calls-to-action you want included. For example, Twilio SendGrid uses AI for recommendations on deliverability, while Optimove uses data and insights to offer the most relevant information even when emails may linger unread in inboxes, or Seventh Sense, which works with HubSpot and Marketo to personalise email delivery times for everything who has signed up to receive your communication.
AI is being integrated into most popular email services, such as Mailchimp. And it can also be used to improve cold emails to potential customers at scale.
With new AI services launching every day, there are lots of potentially useful tools for freelancers aiming to scale up their business. Whether that’s directly within your work, or potentially taking care of other tasks in your life to reduce your stress and commitments.
You could use AI to work with potential clients to refine their briefs before you get involved, meaning that you have clear requirements before you start working with them. Or to transcribe and summarise all of your conversations on Slack to make sure you have a permanent record.
We’ve mentioned AI email tools for marketing and client contacts, but a service like SaneBox can help you to manage your Inbox, and there are solutions for pretty much every administrative task in your business and home life. From invoicing to researching a topic, transcribing virtual meetings, and removing background noise, planning travel itineraries, summarising longer articles for quicker reading or optimising your CV, if there’s a task which relies on data, there’s going to be an AI service available for it.
here are many predictions about the potential impact of AI on the global workforce, including the self-employed. A lot will depend on whether artificial intelligence continues to progress despite the challenges presented by legal issues, potential regulations, and technical limitations.
At present, it’s difficult to see AI completely replacing freelancers, as many of the potential issues it can cause may be business critical. Simply publishing content without any checks could mean irreparable damage if it leads to an audience damaging their health, finances, or other similarly important aspects of their lives.
And the issue of copyright could necessitate human creativity if a business wishes to retain ownership over their IP and published work.
But the freelance workforce is likely to be reduced by the increased use of AI by both clients, and those working for them. The increases in productivity simply mean less people will be needed to produce a similar or increased output. And that means those freelancers who are familiar with the latest AI tools and services will be able to scale their business in size or revenue, while those who tend to offer services based on simple and repetitive tasks are most likely to suffer as a result.
Despite much talk from proponents of AI about the potential new opportunities it offers for careers, so far, the only concrete example has been AI Prompt Engineers, but others may emerge in the future. And as the self-employed are the most likely to continue reskilling and upskilling throughout their working lives, there’s a reason to be more optimistic for freelancers than the majority of the employed workforce.
Looking for more guidance?
Have you ever had a business idea that needed significant capital to start? Angel investors offer an alternative to business loans or venture capital (VC) funding...
In this guide, we run through the key things you need to know about IR35 as a self-employed professional, including the difference between 'inside' and 'outside' ...
Even the most resilient self-employed professionals can be thrown by unexpected challenges, including power cuts, internet outages, office thefts and more. Find o...