Ethics of AI: 10 Main Issues and Code Examples

What are AI ethics?

AI ethics is a set of guiding principles to create AI tools that don’t inadvertently harm or mislead others. 

Key areas within AI ethics include:

• Avoiding AI bias: AI needs huge sets of data to perform, and that data comes from humans. While the algorithms themselves are usually impartial, the information they rely on might be skewed. For example, it can sideline minorities or underrepresented groups, leading to unfair outcomes. It’s essential to develop AI technologies that are inclusive and fair.
• AI and privacy: With AI’s heavy reliance on data, especially in fields like social media and email marketing, there’s a growing concern about user privacy. Basically, AI ethics help ensure that users know their rights, responsibilities and can protect their privacy.
• Managing AI environmental impact: The technological change brought about by AI isn’t just digital — AI models consume massive amounts of actual energy and bring a severe environmental impact. This demands both global attention and responsible actions from businesses around the world.
• Accountability: With the rapid pace of technological change, who’s responsible when AI goes awry? The debate on accountability in the AI realm is crucial and ongoing.
• Proper oversight mechanisms: Implementing human-in-the-loop and human-on-the-loop strategies can help in maintaining control over AI systems. These mechanisms involve human intervention to oversee AI operations, ensuring the systems function ethically and correctly. 

AI is affecting everything from email marketing to content creation, and understanding its ethical dimensions has never been more important.

Why are AI ethics important?

The importance of AI ethics is explained by a simple fact that in three years, AI has become an essential part of people’s lives and affects every aspect of it. While it can be helpful in many ways, it can also bring huge negative consequences. Misinformation, privacy concerns, or even criminal acts — scammers are already using it to imitate humans and gain access to both personal information and bank accounts.

Of course, it’s not all about malice. A lot of AI risks have to do with a lack of understanding or skills to use it. That is why your business should come up with AI ethics rules and follow them. For example:

• Reputation and trust: Ethical missteps can tarnish a brand’s image. Your business can inadvertently harm a massive group of people — just because you used the wrong data set to teach an AI customer support bot. 
• Legal and regulatory concerns:  With AI’s mainstreaming, global governments are formulating regulations for its ethical deployment. Ignoring them can lead to legal complications and hefty penalties.
• Genuine business value: Ethical AI can benefit your business — and become a unique selling proposition in today’s data-centric world. But for that, your product needs to be privacy-focused. 
• Societal impact: AI’s potential societal impact is vast. Ethical considerations guide it towards inclusivity and fairness rather than division and bias.
• Future-proofing: As AI tools become more prevalent, businesses prioritizing ethics will navigate future challenges and shifts with ease.

In essence, AI ethics is a pressing concern that every marketer, business owner, and even the general public should be aware of.

10 main ethical issues of AI today

As we approach significant technological change, it’s essential to weigh the risks and rewards. A Pew Research study found that 79% of experts expressed concerns about the future of AI, touching on potential harms to human development, knowledge, and rights. Let’s break down the ten main ethical issues of AI today, spiced up with real-life examples.

Bias and discrimination

As we’ve mentioned, AI can be biased — that’s because humans are. But with AI tools consequences become even more vast and dire. The Berkeley Haas Center for Equity, Gender, and Leadership revealed that out of 133 AI systems analyzed from 1988 to the present day, 44.2% exhibited gender bias, and a significant 25.7% showed both gender and racial bias.

That set a precedent with New York City in 2023: now, employers are prohibited from using AI to screen candidates unless the technology has undergone a “bias audit” in the year preceding its usage.

Lack of transparency

AI systems, especially in marketing, often operate as “black boxes”, making decisions that even their creators can’t always explain. This creates a lack of transparency, and an unsettling one.  For instance, if an AI in email marketing decides to target a particular demographic without clear reasoning? It’s not only the brand reputation that might be tarnished, but also a group of people badly hurt by it. That is exactly why it needs human oversight — and a final say by humans in any matter. 

Invasion of privacy and misuse of surveillance

AI gives both governments and businesses an unprecedented amount of access to our personal lives. Artificial intelligence can skim through your social media accounts, personal data, and internet history. It can predict your behavior or, for example, decide if you get a dream job. 

All of this means that the use of AI needs to be heavily regulated — and businesses need to comply with the rules. Thus, information won’t be misused (or at least not as badly as it could be). 

While the discussion about global AI safety measures is still ongoing, some actions have already been taken. In the US, AI guidelines are now more relaxed thanks to the Executive Order for Removing Barriers to American Leadership in AI (“Removing Barriers EO”) that was issued by President Donald Trump in January 2025. This rescinds the “Order for the Safe, Secure, and Trustworthy Development and Use of AI” from the previous administration. As you can get from the names, the latest order is less concerned with the ethics of AI. At the same time, several AI companies, like Adobe, Google, and Meta, are committed (at least on paper) to the transparent and safe development of AI technology. 

Meanwhile, the European Union passed the first AI Act that bans some AI applications, including cognitive behavioural manipulation, social scoring, biometric identification and categorisation, and real-time and remote biometric identification systems, such as facial recognition in public spaces.

Lack of accountability

One of the most complicated AI ethics problems is accountability. Let’s say, something went wrong — do you blame a developer who created the algorithm, the user who fed the algorithm with invalid or biased data, or AI itself?

This is already a recognised issue: in the US, the Algorithmic Accountability Act was proposed in 2023. It mandates that AI companies need to conduct an impact assessment to identify potential risks to customers and users. They also need to be transparent about how algorithms work and how they make decisions. This can protect regular users and help them make informed decisions. As of 2025, this act was introduced in both the House and Senate, but it has not yet been passed into law.

Copyright questions

Another topic of debate surrounding AI is the use of protected works (illustrations, books, etc.) to train the systems and generate text and images. The AI and copyright questions are at the intersection of law, ethics, and philosophy, which makes them especially complicated to navigate. A lot of the effort here comes from businesses. 

In 2025, Disney and Universal sued Midjourney because of the alleged copyright infringement. This might mark the start of changes in how AI image generators will operate in the future. It also challenges marketers who use AI to produce any kind of design and illustrations.

Economic and job displacement

AI is already part of the workflow and routine in many fields. So the question is — can it replace humans? So far, the answer is no. Still, artificial intelligence can make it harder for people to find a new job. For one, a lot of focus will be placed on creative tasks and the ability to oversee AI’s work, and a lot of people around the world lack the necessary skills to do that.

Misinformation and manipulation

Ever stumbled upon a video of Tom Cruise goofing around, only to realize it was all smoke and mirrors? Those kinds of videos are deepfakes — hyper-realistic videos of individuals saying or doing things they never actually did, created by AI. Of course, some clips are better than others. But what can you do with plain text? 

In a startling case that shook the internet, two US lawyers and their firm were slapped with a $5,000 fine for leveraging ChatGPT to create and submit fake court citations. This AI tool went as far as inventing six non-existent legal cases. Imagine every court case filled with non-existent laws and examples — and a precedent this might set.

Safety concerns

AI technology can sometimes remind a toddler — unpredictable and needing constant supervision. While it has the power to generate groundbreaking tools and solutions, it can also act in unforeseen ways if not properly managed. Consider the global buzz around autonomous vehicles. Can we fully trust AI to take the wheel in critical situations, or should there always be a human overseeing its actions to ensure safety?

Recall the tragic incident in Arizona during Uber’s self-driving experiment that led to fatal consequences for an innocent pedestrian. Similar concerns have been raised with Tesla’s autopilot feature. These incidents bring us to a pressing question in the business and technology world: How can we strike the right balance between leveraging AI’s potential and maintaining safety?

Ethical treatment of AI

While a lot of our questions regard the treatment of humans… how should we treat AI itself? 

As AI systems evolve, there’s a growing debate over their rights. Are they just sophisticated tools or entities deserving of respect? While this is a question of ethics, it has practical implications, too. For example, the quality of AI’s response in some cases depends on how polite and articulate you are.

Over-reliance on AI

Relying solely on AI, especially in critical areas, can be a recipe for disaster. We’ve already mentioned a criminal case where over-reliance on AI led to penalties, but imagine your doctor relying on ChatGPT for advice and never evaluating its accuracy. Here is a different scenario — a politician can support his or her point with made-up statistics. AI definitely needs human oversight. 

So, as we embrace AI, let’s navigate its ethical landscape with care. After all, it’s not just about what AI can do, but what it should do. One thing that can help with that is an AI code of ethics.

What is an AI code of ethics?

AI code of ethics is a set of guidelines that companies and organizations use to ensure that their AI tools are used responsibly and ethically. Ideally, every company that uses AU should have this rulebook. This way, they don’t compromise on human values while chasing better and more efficient technologies. 

Still, creating AI ethics code is a process. And according to HBR, here’s how you can get started:

1. Gather the stakeholders: Get diverse minds on board to avoid tunnel vision. Make sure everyone is represented — those who get both computer science and human science as well.
2. Set your standards: Clearly define what ethical AI means for your organization. This isn’t a one-size-fits-all scenario — your company’s mission and values play a crucial role.
3. Spot the gaps: Set up your goals for AI and map out your company’s journey to achieve them.
4. Get to the root: Dive deep into potential issues and find practical solutions.

And hey, if you’re looking for some inspiration, UNESCO set a global standard with their AI ethics framework in 2021. Now that’s a solid foundation to build upon.

Examples of AI codes of ethics

Let’s take a deeper dive into the AI codes of ethics from some big players — IBM, BCG, Bosch, and the European Commission. We’ll try to spot the differences and similarities. From grasping the basics to exploring the unique features, we are set for an exciting journey.

BCG (Boston Consulting Group)

Purpose: BCG’s AI Code of Conduct is designed to guide the responsible development and use of AI. It helps BCG in advising clients on how they can integrate AI into their businesses ethically while benefiting from it.

Key principles:

• Human benefit: AI should be used for the benefit of all and should not harm humanity.
• Fairness: AI should be unbiased and should not perpetuate existing biases.
• Transparency: The workings of AI should be transparent and understandable.
• Reliability: AI should be reliable and safe.
• Privacy: AI should respect privacy and data protection rights.
• Accountability: There should be clear accountability for AI’s decisions and actions.

Bosch

Purpose: Bosch’s ethical guidelines for AI is focused on responsibly and respect to human dignity. The company’s global partners have adopted these guidelines to foster AI technologies that prioritize human well-being and ethics.

Key principles:

• Beneficence: AI should be used for the good of humanity.
• Non-maleficence: AI should not harm humans.
• Non-autonomy: Humans should always be in control of AI.
• Justice: AI should be used fairly and should not discriminate.
• Explicability: AI’s decisions should be transparent and explainable.

IBM

Purpose: IBM’s Everyday Ethics for Artificial Intelligence provides a set of guidelines for the ethical development and use of AI. Thousands of companies using IBM’s AI products naturally inherit those guidelines and make sure that AI utilization is both ethical and transparent.

Key principles:

• Accountability: AI developers and users should be accountable for their AI systems.
• Value alignment: AI should be aligned with human values and ethics.
• Explainability: AI’s decisions should be understandable to humans.
• User data rights: AI should respect user data rights and privacy.
• Fairness and non-discrimination: AI should be unbiased and should not discriminate.
• Transparency: AI’s workings should be transparent.
• Social and environmental well-being: AI should be used in a way that benefits society and the environment.
• Safety and security: AI should be safe and secure.

European Commission

Purpose: The High-Level Expert Group on AI presented these guidelines to ensure the development of trustworthy artificial intelligence. The Ethics Guidelines for Trustworthy Artificial Intelligence were formulated after an open consultation that received more than 500 comments. These guidelines are utilized by AI developers globally to create AI systems that are ethical and reliable.

Key principles:

• Lawfulness: AI should respect all applicable laws and regulations.
• Ethical behavior: AI should respect ethical principles and values.
• Human agency and oversight: AI should empower humans, allowing them to make informed decisions. Proper oversight mechanisms, such as human-in-the-loop and human-on-the-loop, should be in place.
• Technical robustness and safety: AI should be resilient, secure, safe, accurate, reliable, and reproducible.
• Privacy and data governance: AI should respect privacy and data protection rights. It should also ensure data quality, integrity, and legitimate access.
• Transparency: AI systems and their decisions should be transparent. Humans should be aware of their interactions with AI and understand its capabilities and limitations.
• Diversity, non-discrimination, and fairness: AI should avoid biases and foster diversity. It should be accessible to all and involve stakeholders throughout its lifecycle.
• Societal and environmental well-being: AI should benefit all humans and consider its environmental impact. It should be sustainable and environmentally friendly.
• Accountability: AI should have mechanisms for responsibility and accountability. This includes the ability to check and easily address any issues.

How to make AI more ethical

Does more ethical AI rely only on regulation and big businesses? Of course not. Both developers and users can responsibly use artificial intelligence tools to make sure that the results are safe and unbiased. Here are a few simple steps you can take to use AI or develop AI tools ethically:

• Use diverse datasets for training. Make sure your AI tool or platform can represent different perspectives and use different lenses. It will also help to check for biases in the AI system, so you’ll know that no groups are discriminated against. 
• Always check and audit your AI tool. This is to ensure the first principle is met. 
• Be transparent and open. If you develop any kind of AI tool, clearly define its abilities and limitations, plus notify your users about the potential impact. It will also help to communicate the exact way your AI is working and “thinking”.
• Make the roles clear. This is the first step towards accountability: make sure users know who is developing, monitoring, and overseeing AI. 
• Check for relevant regulations and collect only necessary information. Of course, you will need to ask for permission first. 

If you are looking for inspiration, Salesforce’s curated list of ethics in AI research papers, updated regularly, offers a treasure trove of insights. You can also use resources from pioneers in promoting ethical conduct in AI, such as AlgorithmWatch, AI Now Institute, DARPA, CHAI, and NASCAI.

Final thoughts

Hopefully, navigating the intricate landscape of AI ethics hasn’t been boring. Here are a few key takeaways:

1. Use AI responsibly: While AI may transform many sectors, including email marketing, it’s essential to use this tool with care. 
2. Don’t forget about human oversight: AI, for all its brilliance, requires human guidance. So if you worry whether AI tools replace human copywriters, don’t be — copywriters are still here to oversee the ethics of it all. 
3. Set ethical boundaries: Establishing a robust AI code of ethics isn’t just good practice, but a necessity. Thinks of rules and codes as a compass. 
4. Stay ahead of the curve: The AI landscape is dynamic. For marketers, especially those leveraging new technologies in marketing, it’s crucial to stay informed and adapt.

As AI becomes a bigger part of everyone’s life and your business in particular, its ethical use will not only earn people’s trust but also spark more creative ideas.