Do you turn to ChatGPT to help you flesh out a monthly report? Or perhaps you’ve become reliant on Claude.ai to help you construct company-wide update emails in a matter of seconds, instead of painstakingly labouring over your sentence structure for hours?
Despite 93 per cent of employees and business leaders having concerns about implementing AI in the workplace in a more formal and structured capacity, the reality is that UK workers are increasingly using generative AI to help them complete daily tasks.
In fact, as many as 18 million people aged between 16 and 75 have used generative AI in the UK, according to research from Deloitte, which equates to a third of the population.
5 data scientist roles hiring across the UK
- Data Scientist, NatWest, Edinburgh
- Data Steward, Brakes, Ashford
- Sr. Data Scientist, EU S/C, Supply Chain Science, Amazon TA, London
- Data Engineer, London, Publicis Media, UK
- Master Data Administrator, KP Snacks, Slough
So, why the disconnect between those who favour a bring your own AI (BYOAI) approach and senior leaders who are reluctant to adopt proper AI guidelines?
“Employees are moving faster than their employers when it comes to adopting GenAI to transform how they work. While workers are signalling that GenAI can boost their output and save them time, many employees may not be supported, encouraged, or explicitly endorsed to use the technology by their organisation,” offers Paul Lee, the partner and head of technology, media and telecommunications research at Deloitte.
This sentiment is echoed by Daniel Pell, vice president and country manager, UKI at Workday. “It’s important that we recognise AI upskilling is not only a technical problem. It’s also cultural,” he says.
“In fact, one of the biggest barriers to AI adoption today is trust. This trust gap can only be overcome through clear communication of AI’s use alongside strategic implementation in areas that benefit employees and businesses.”
One of the major ways generative AI boosts productivity and output is by the vast amounts of data it can process in a matter of seconds so it makes sense that data scientists in particular have the added advantage of harnessing their extensive knowledge and methodologies with all the benefits generative AI has to offer.
With this in mind, discover 10 useful prompts that data scientists can use in their everyday work to clarify concepts, weigh up options or simply generate code for common tasks.
Prompt 1: ”Explain [specific data science concept or algorithm] in simple terms”
This can be used to quickly clarify complex concepts like neural networks, random forests, or gradient descent.
Prompt 2: “What are the pros and cons of using [specific model or algorithm] for [specific task]?”
This prompt is helpful for weighing options when deciding on a model or algorithm for a particular use case.
Prompt 3: “Generate a Python script for [data science task], such as data cleaning, EDA, or feature engineering.”
This prompt will help you save time by quickly generating code for common tasks.
Prompt 4: “Provide a brief overview of the latest research or trends in [specific area, e.g., machine learning, NLP, etc.]”
Useful for staying updated on recent developments in the field, you can use this prompt to receive a brief rundown of updates. You can also request that you receive the information in bullet point form or limit the word count to keep it as succinct as possible.
Prompt 5: “How do I interpret the output of [specific model or statistical test]?”
This prompt can help in deciphering the results of models or tests, like logistic regression coefficients or p-values.
Prompt 6: “Suggest ways to improve the performance of my model”
If you’re looking for tips on hyperparameter tuning, feature selection, or model selection to enhance accuracy, this prompt will offer a variety of options.
Prompt 7: “What are the best practices for handling missing data in [specific dataset]?”
This prompt provides strategies for dealing with missing values, such as imputation methods or data exclusion.
Prompt 8: “Help me understand the implications of [specific metric or KPI] in my analysis”
If you need assistance in interpreting metrics like AUC-ROC, precision, recall, or R-squared in the context of the analysis, this prompt will guide you in the right direction.
Prompt 9: “Draft a summary of my data analysis results for a non-technical audience”
This prompt is extremely useful for translating technical findings into understandable insights for stakeholders.
Prompt 10: “What are common pitfalls to avoid when working with [specific data type or problem, e.g., time series, imbalanced data]?”
If you want to highlight potential issues and how to address them, this prompt will ensure you conduct a more robust analysis.
Read more
Writing your company’s own ChatGPT policy – Giving your staff unfettered access to ChatGPT could have disastrous results, especially when it comes to unwittingly releasing confidential data. Here are some pointers as to how to write your company’s own generative AI policy
ChatGPT won’t kill Google — it makes search more important – Here’s how the business and societal roles of Google and the wider search engine space is set to evolve alongside generative AI tools like ChatGPT
What ChatGPT means for developers – Will ChatGPT replace software developers? Or will it make their jobs more enjoyable by doing the heavy lifting when it comes to writing boilerplate code, debugging and testing? David Adams speaks to developers already using ChatGPT
