Are you already using or considering using AI (artificial intelligence) in your business; whether in the form of product delivery to your customers or clients, or to help in the day-to-day administration of your business? AI is fast becoming an accessible and often cost-effective tool to improve efficiency in areas such as customer service, data analysis, content writing, and automation of processes. However, the use of AI is not without ethical and legal issues and understanding these can help you mitigate risks for your business.
‘Currently, the UK does not have any AI-specific legislation, instead the use of AI is regulated under existing legal frameworks in respect of its application in a specific context, and businesses must ensure they remain compliant as they embrace the use of AI within their business,’ explains Charles Day, a Trainee Solicitor in our corporate and commercial team.
Since AI has so many potential use cases, your business strategy will help you identify where you want to use AI and for what purpose. Once you have, we can then help you identify the relevant legal considerations for a specific implementation. For example, if you are wanting to use AI for data analysis, careful consideration and advice should be sought around data protection laws, maintaining confidentiality and protecting trade secrets. Similarly, if AI is important for you in generating reports or marketing content, consideration would need to be given to areas such as intellectual property, disclosure, accuracy and consumer legislation.
Charles Day highlights a few key considerations in regard to commercial activities, but a tailored approach will be needed for each business and the particular circumstances, given the breadth of possible applications which AI may have for your business and its intersection with different legal considerations. The use of AI in recruitment and employment is not covered in this article.
Data protection and privacy
Businesses are required to handle and process personal data with transparency, minimal processing and accountability. If you are using AI for data analysis, this may require the input of sensitive data or related information. AI relies on large amounts of data to optimise its effectiveness and this in turn raises risks of data breaches for your company. This, therefore, increases the need to put in place proper policies, documentation or protocols to ensure compliance.
Intellectual property
If you are using AI in your business to help you by generating images or materials, then one big question to consider and answer is – who owns the intellectual property rights to that content and do you need to action anything to protect your business? Is the content created by a third party, and have they licensed it to the AI platform you are using? Is it the property of the AI platform itself, and are you familiar with the terms and conditions around use of that material or image? To avoid inadvertently infringing intellectual property rights, it is prudent to first check the terms and conditions of the AI platform and to seek legal advice about further steps you may need to take.
Risks associated with AI outputs
If, for example, you wish to use AI to write product descriptions, there is an inherent risk with AI in that it is not always accurate. In the case of using AI for product descriptions, if critical information is missing about the product or appropriate disclosures were omitted, this could put your business at risk and out of compliance with legislation such as the Consumer Rights Act 2015, or any industry specific regulation, where businesses are under an obligation to provide accurate information. Therefore, human fact checking or oversight is still critical.
Another important reason to oversee AI outputs would be to monitor any cultural or ethical sensitivities the AI generated content may have missed. For example, knowing that the AI has been trained on, and that outputs are pulled from, big data, the generated content will be biased and may contain wording, phrases or even opinions that may cause offence, be culturally inappropriate or out of alignment with the values of your business. In this instance, the need for the human element would be crucial to preserving the ‘voice’ and credibility of your business with your customers or clients. In many instances AI generated content can be quickly and easily identified – the use of AI itself could damage your brand image and alienate people, even where the content does not explicitly fall short of regulatory requirements, cause offence, contradict your values or contain any other errors or inaccuracies.
Transparency and accuracy
Also related to AI outputs, further reputational risk considerations may come into play if inaccuracies from the use of AI lead to misrepresentation or similar liability.
For example, if a customer engaged you to write original content, such as a report or feature article, but you used AI to assist you. They would be expecting you or your employees to have produced the work and may feel they have been mis-sold if they then discover the piece was AI generated. Failure to disclose the use of AI in your business can lead to very serious ramifications.
It is worth noting that under the EU AI Act, there are strict parameters around transparency and disclosure of the use of AI. While the UK does not have AI-specific legislation dealing with labelling requirements yet, it is advisable that, in addition to addressing legal issues with AI integration, protocols are put in place around disclosure and transparency to best protect your business and its reputation.
How we can help
AI integration into your business clearly has a wide scope to touch on many discrete legal and commercial considerations and whether the particular concern for your business relates to data protection, intellectual property, marketing uses, or even cost reduction and efficiencies, it is important to undertake due diligence.
Our commercial solicitors would always consider the particular nature of your business and the specific circumstances around your planned use of AI.
Our advice would cover a range of issues which could include:
- identifying which legislation will be engaged and impact you by your use of AI and how to stay compliant;
- reviewing the terms and conditions of the AI platform to ascertain any restrictions you ought to be mindful of when using AI generated material for your purposes;
- reviewing and updating your business documentation, such as terms and conditions or contracts with clients and customers. Does the use of AI then require you to introduce AI specific clauses in your contracts? Drafting of liability clauses to protect you against claims from the use of AI, disclaimers and caveats around the use of AI and adjusting your privacy and data processing policies will require an overview of what documentation would need amending as a direct result of integration of AI within your business; and
- identifying further good practice measures which may be needed. Depending on the extent of AI integration you are planning, it may be necessary to consider the drafting of an AI policy to include, for example, information around your values, your ethical use of AI and guidelines you intend on following to ensure continued responsible use of AI.
This is not an exhaustive list but aims to give you an overview of steps you may need to consider to ensure your house is in order with the use of AI, just as you would with any other corporate and commercial documentation.
Our team of commercial solicitors can help you navigate this fast-evolving area of business law to ensure your business can benefit from AI integration while remaining compliant now and for the future, being pro-active and following best practice.
For further information, please contact Charles Day in the corporate and commercial team on 01872 241414 or email CDay@nalders.co.uk. Nalders has offices across Cornwall in Truro, Falmouth, St. Austell, Camborne, Helston, Newquay and Penzance.
This article is for general information only and does not constitute legal or professional advice. Please note that the law may have changed since this article was published.