Consider you’re a busy parent looking for ways to simplify your life. Putting your weekly schedule or grocery list into a generative AI chatbot like ChatGPT could be a saving grace, assisting in tasks like weekly meal planning or even helping your child with homework. This kind of automation via generative AI shows how it can be beneficial in streamlining a person’s daily tasks and managing their affairs.

But how dangerous is AI for the future of work?

There has been a long-standing fear that machines will eventually replace human professions, alarming workers about the future of employment. Reports suggest that up to 300 million jobs could be replaced by AI, generating looming uncertainty over workforce dynamics and automation.

So, every working person’s question is: Will AI take away my dream job? In short, the answer is likely yes, but it depends on what your dream job entails.

However, it is worth highlighting the positive ways AI augments certain workforces. In high-stress or tedious positions, automation can enable workers to focus on more meaningful and high-value activities while relieving some job-related burdens.

While it’s undeniable that AI will continue to reshape industries, there is substantial reason to believe it won’t take over every job. Yes, AI does offer capabilities that lend a hand to productivity. However, its abilities have inherent limitations, especially falling short when replicating human intuition, empathy, and creativity.

We also can’t ignore that AI is limited to the information it’s trained on. OpenAI touts ChatGPT’s ability to pass numerous standardized tests, but did the model genuinely understand the tests, or was it just trained to reproduce the correct answers? AI machines are designed to “think within the box,” meaning that they can only function within the parameters of their given data, lacking a true creative or analytical nature that many professions require.

Related: AI Is Taking Over These Freelancing Jobs the Most: Report

AI’s thinking gap

AI has been a game-changer for law by automating repetitive tasks, but integrating these technologies comes with challenges. The Law Society of British Columbia, for example, has warned lawyers about using AI automation to prepare legal documents, highlighting the potential ethical implications, biases, and glaring inaccuracies in materials produced by AI.

Likewise, the world of architecture is no stranger to technological advancements, with architects always looking for ways to harness innovation to improve their designs. But when AI tools can generate blueprints with relative ease, the fear of architects becoming redundant remains. While an AI model can create and analyze design options or predict how people can use a space, it cannot replace the vision and intuition an architect brings to their craft.

Healthcare is another domain where complexities of human emotions and behaviors necessitate the expertise of trained professionals who offer empathy and deep understanding. AI can undoubtedly assist with administrative tasks like analyzing data or offering high-level insights, but it lacks the essential human qualities needed for effective counseling.

AI enthusiasts in healthcare have long envisioned a future where doctors are aided by sophisticated algorithms, providing valuable suggestions to enhance patient care. However, despite a high level of optimism about physicians benefitting from AI, regulatory uncertainty and concerns about patient experience have slowed adoption. A recent study found that ChatGPT has a more than 80% diagnostic error rate when using AI in pediatric case diagnosis — an unacceptable figure that indicates no doctor could ever rely on its output now. But could the pieces of this puzzle come together to build physician and public confidence in AI?

Related: Report: AI Will Take More Jobs Away from Women Than Men

Building trust with automation

While successful AI implementation will likely positively impact patient outcomes, the consequences of errors would be severe. Even if a physician doesn’t take a potentially harmful AI suggestion at face value, they will end up spending valuable time and resources rectifying the inaccuracies.

AI’s opaque decision-making processes also pose challenges for clinicians who cannot understand the rationale behind automated recommendations. This lack of transparency not only undermines trust in AI systems but complicates the process of integrating them into current healthcare infrastructures.

That being said, some startups are working to make AI accountable in the field by rooting it in reality. Kahun, for instance, developed an evidence-based clinical reasoning engine for doctors that provides clinical decision support using explainable AI. It reasons like a physician and allows transparency in its output by providing evidence-based clinical insights cited in peer-reviewed medical literature alongside patient-specific information.

Datapeople is another example of an AI-assisted tool designed to support professionals in the workforce. This program assists hiring teams in making recruitment decisions grounded in scientific principles and fairness.

Suppose professionals embraced the benefits of AI rather than fearing its encroachment on their jobs. In that case, they may be more inclined to integrate these tools into their practice, leading to potential improvements in the quality and efficiency of their work.

Of course, not all jobs are safe from AI. But the truth is that many roles will always require human intervention, particularly those involving decision-making, emotional intelligence, and creativity. However, combining AI’s strengths with human capabilities allows for an intertwined approach to problem-solving and innovation. By leveraging AI’s efficiency with human intuition, we can continue to use automation to optimize tasks and address complex challenges more effectively.