Cognilytica’s Prompt Engineering Best Practices Guide: Using Custom Instructions (Part 3 of 6)

Cognilytica’s Prompt Engineering Best Practices Guide: Using Custom Instructions (Part 3 of 6)

This week we share part 3 of our 6 part newsletter on Prompt Engineering Best Practices. Make sure to check out the first part here and second part here if you haven’t already!

Prompting Best Practice #3: Use Custom Instructions

Getting the LLM to do your bidding is always the hardest part. Prompt engineering puts the “Art” in artificial intelligence. Perhaps writing effective prompts should be a standard course in middle school and high school English classes? That’s another topic for another day.

One of the ways of coaxing these LLM systems to providing the desired output is to provide greater context and information that can be applied to one or more prompts. Not only does this help the LLM system provide better focused output, but it can also help to avoid continuously repeating yourself in prompts around aspects of tone, format of output, etc.

OpenAI’s ChatGPT in particular allows you to provide settings that tailor the responses to match your tone and desired output goals as a “default” that can be applied to all future prompts. Many have reported that using custom instructions has resulted in much more accurate, precise, consistent, and predictable results.?

You can provide these custom instructions through the API, or find the Custom Instructions in the “Customize ChatGPT” settings:

In ChatGPT, Custom instructions are provided by answering two questions asked in settings that get sent along with your prompts:

  • What would you like ChatGPT to know about you to provide better responses?
  • How would you like ChatGPT to respond?

These custom instructions then will apply to all future chat prompt sessions (not previous or existing ones), allowing you to make somewhat permanent settings that don’t have to be constantly reset. Custom Prompt instructions are short and generally limited to about 1500 characters, so keep it precise and concise.

What would you like ChatGPT to know about you to provide better responses?

This section focuses on ensuring that responses match your personal style, tone, expertise level, and background that should be reflected in the prompt responses. The best way to use this is to craft a unique voice? and ensure that the system doesn’t forget your settings between different chat prompt sessions.?

For example, you can tell the LLM that “I’m a professional working with project managers and other business and technical implementers of AI, so assume that my audience will be somewhat knowledgeable on the topic of AI and generative AI, and very knowledgeable on topics of project management”

How would you like ChatGPT to respond?

With custom responses, you can ensure consistent results with custom prompt responses, and set expectations for the type of output, your goals, or even format of the output

For example, you can tell the LLM to “Ask clarifying questions that will help you generate a better response before starting. If the questions will not change your output, then respond with the output”.

You can also use Custom Prompt instructions to specify negative prompts: those things you never want to see in any LLM responses.

For example, “Never use any first-person pronouns in your responses. Always provide reference sources.”

? Stay subscribed for the part 4 in our series next week where we explore another best practice in prompt engineering:?

? NEXT UP IN PART FOUR: We’ll be discussing Using Plugins to enhance the LLM's capabilities beyond its base functions.


?? In Case you Missed it… Insights from Cognilytica Podcasts & Content

  • ?? {Featured Forbes Article} What Are The Real Costs Of AI Projects? When you're trying to run and manage AI projects, there's a lot that needs to be considered in terms of determining project budgets and cost. What are the real costs of running AI projects?? [Read here ]
  • ?? {Featured Forbes Article} Are Increasingly Competitive AI Markets Pushing Users To Open Source? AI is making its way into all types of products and services that we’re using increasingly on a daily basis. What does that mean for a new form of vendor lock-in? [Read More ]
  • ?? {Featured Podcast} AI’s impact in the practice of medicine: Interview with Dr. Jag Singh, Mass General Brigham? - AI’s impact is helping to reshape the practice of medicine both now, and in the future. In this episode of the AI Today podcast we interview Dr. Jag Singh. He is a Professor of Medicine at Harvard Medical School, focusing on the application of AI in the practice of medicine. He is also a practicing cardiologist at Mass General Brigham (MGB). [Listen here ]
  • ?? {Featured Podcast} How AI will Re-imagine Air Travel: Interview with Bernadette Berger, Alaska Airlines - AI is transforming the aviation industry and passenger experience in some profound ways. In this episode of the AI Today podcast we interview Bernadette Berger, who is Director of Innovation at Alaska Airlines as she shares with us how AI will reimagine air travel and the airport experience. [Listen here ]

?? The Most Important AI Stories from the Past Week You Should Know:

  • {AI Strategic Plans} - Canadian PM announces $2.4 billion for AI-related investments - Canadian Prime Minister Justin Trudeau announced Canada will be setting aside $2.4 billion in its upcoming budget to build capacity in AI. The investment will be divided between a number of measures aimed to advance job growth in the AI and tech industry, but it appears the majority (about $2B) is going into a fund that will provide access to computing capabilities and technical infrastructure. [Read More ]
  • ?? {Autonomous Pattern} - Amazon ditches cashierless checkout system at its grocery stores - The Autonomous Pattern of AI continues to be the hardest to implement. Amazon announced it’s removing its cashierless checkout systems, called Just Walk Out, at Fresh supermarkets in the U.S. Another case of overpromise and underdeliver with AI, and yet again proof that the Autonomous pattern is the hardest pattern of AI. [Read More ]?
  • ?? {AI Regulation} - AI Safety: UK and US sign landmark agreement. The agreement signed on Monday says both countries will work together on developing "robust" methods for evaluating the safety of AI tools and the systems that underpin them. AI safety continues to be a regulatory priority.? [Read More ]
  • ??? {AI Regulation} -? Washington state judge blocks use of AI-enhanced video as evidence in possible first-of-its-kind ruling - A Washington state judge overseeing a triple murder case barred the use of video enhanced by AI as evidence in a ruling. The ruling comes as AI and its uses — including the proliferation of deepfakes on social media and in political campaigns — quickly evolve. Will we see the use (or banning) of AI continue to play out in the courts? [Read More ]
  • ?? {AI Applications} - Insurance companies rely on aerial images to ditch risky properties - Insurance companies are leveraging aerial images of homes to drop properties deemed too risky. AI models scrutinize the images looking for underwriting hazards, such as “damaged roof shingles, yard debris, overhanging tree branches or undeclared swimming pools or trampolines.” Have insurers gone too far with AI-powered surveillance? [Read More ]
  • ?? {Autonomous Pattern} - Apple Lays Off Workers After Canceling Car Project - The autonomous pattern of AI continues to be the hardest pattern to implement.? Apple had axed its decadelong plan to build a self-driving electric car and is laying off over 600 workers. Have we reached peak Autonomous Vehicle? [Read More ]
  • ?? {AI Experience} - AI-aided virtual conversations with WWII vets are latest feature at New Orleans museum - The? National WWII Museum in New Orleans recently opened an interactive exhibit. With the help of AI visitors can hold virtual conversations with 18 veterans of the war. [Read More ]

?? Events and Opportunities to Hear from Cognilytica

  • April 17, 2024: PMI New York City Chapter “How to Run Successful AI Projects and Avoid Failure” Virtual / Online session, 6-7pm EST [Register here ]
  • April 18, 2024: FORUM: Women Making an IMPACT, Empowering Women in AI - virtual/ webinar, 1pm EST [Register here ]
  • April 23, 2024: PMI Tulsa, OK Chapter: “The Hidden Secret of AI Success: A PMI Approach” Virtual / Online session 12:00-1:00pm CST [Register here ]
  • April 29, 2024: PMI Saudi Arabia Chapter: “How to Run Successful AI Projects and Avoid Failure”, Virtual / Online session -? 7PM-8PM Saudi Time (10AM-11AM Eastern Time) [Register here ]
  • May 2, 2024: PMI Austin, TX Chapter: “The Future is Now:? Essential AI Skills Every Project Manager Must Have”, Professional Development Day 2024, In-person presentation, 7:30am - 16:00pm CT ?[Register here ]
  • May 7, 2024: PMI Delaware Valley Chapter: “How to Run Successful AI Projects and Avoid Failure”, Virtual / Online session -? 6:30PM-8PM Eastern Time [Registration opening soon]
  • May 8, 2024: PMI Southwest Ohio Chapter: “How to Run Successful AI Projects and Avoid Failure”, Virtual / Online session, 6:30-7:30pm Eastern Time [Register here ]
  • May 15, 2024: Designing Maryland's AI Future Summit, 8:30am-4:30pm, Fort Washington, MD [Registration opening soon]
  • May 21, 2024: PMI New Jersey Chapter: “What you need to know, today, to successfully run and manage AI projects”, Virtual / Online session -? 6:30PM-8PM Eastern Time [Registration opening soon]
  • July 10, 2024: PMI Madrid, Spain Chapter: “Best Practice Methods for Successful AI Projects”, Virtual / Online session - 7:00 - 8:00 PM Spanish time? [Registration opening soon]
  • July 10th, 2024: PMI San Francisco Chapter: “Successful approaches to running AI Projects - Avoiding the Top Reasons why AI Projects Fail”, Virtual / Online session -? 6-7:30 pm PST [Registration opening soon]
  • July 17, 2024: PMI Northern Alberta Chapter: “The hidden secret of AI success: a PM approach” Virtual / Online session? 6-8pm MT [Registration opening soon]

??? Bring Cognilytica Thought Leadership to Your Organization or Event

? Bring the Power and Excitement of Cognilytica to your Organization or Event!

Cognilytica’s inspiring, thoughtful, and engaging analysts deliver speaking engagements, workshops, panel participation and moderation, podcast interviews, and other engagements informed by real-world experience and thought leadership.?

Engage our analysts as keynote speakers, panel moderators and participants, workshop facilitators, podcast hosts or guests, and as experts for field marketing events.

?? [Engage with Cognilytica for your next Event, Panel, Keynote, or Activity ]

?? Move Forward with AI Best Practices - Training & Certification

Cognilytica’s AI best practices & Trustworthy training and certification continues to be in high demand. Haven’t yet enrolled in a certification or training? What’s holding you back?

  • ?? FREE Intro to CPMAI - Learn how to Run AI Projects with Best Practices. The CPMAI methodology is the industry’s best practice for AI & ML projects. Take this free, introductory course offered as a self-paced, entirely virtual offering. [Enroll now ]
  • ?? CPMAI v7 - Get Certified with comprehensive AI & ML Project Management Training. Includes: AI Fundamentals, AI Applications, Managing Data for AI, Data Preparation for AI, ML Algorithms, Generative AI, CPMAI Methodology, and Trustworthy AI. [Enroll now ]
  • ?? CPMAI+ Plus v7 - Greater Depth: Enhances CPMAI with RPA, Big Data, Data Science. Includes: All CPMAI Training Content, including AI & ML Fundamentals, CPMAI Methodology, plus Fundamentals of Big Data, Big Data platforms, Foundations of Data Science, Foundations & Applications of Robotic Process Automation (RPA), Big Data Engineering, Security & Governance, and more! [Enroll now ]
  • ?? CPMAI+E v7 - Our most comprehensive training & certification! Adds Ethical & Trustworthy AI to CPMAI+. Includes: All CPMAI+ Plus Training Content, Ethical AI Concepts, Responsible AI Concepts, Transparent AI Concepts, Governed AI Concepts, Explainable AI Concepts, The Trustworthy AI Framework, Putting Trustworthy AI into Practice. [Enroll now ]
  • ?? FREE Intro to Trustworthy AI - What does it take to make AI systems Trustworthy and what can you practically do about it for your projects today? Learn more in this FREE 2 hr self paced mini-course on Trustworthy AI! [Enroll now ]
  • ? Trustworthy AI Framework v3? - Most Comprehensive, Vendor-Neutral Trustworthy AI Training & Certification. Learn how to Build and Run Trustworthy AI Systems. Boost your credentials. Keep Your AI Solutions, Organization, Customers, and Stakeholders Trustworthy. Advance your career. [Enroll now ]

AI Resources

?? Check out our AI & Data Resource List: Dive Deeper! ?

Navigating tone and consistency in AI prompts is key ?? As Aristotle once paraphrased, excellence is a habit. Custom settings are our canvas for creating masterpieces.

回复

要查看或添加评论,请登录

社区洞察

其他会员也浏览了