ChatGPT for local news publishers

In this introductory prompt engineering guide, I’ve included dozens of possible use cases for local newsrooms, a step-by-step breakdown of the anatomy of a ChatGPT prompt, and beginner, intermediate, and advanced prompts for local publishers.

Hay una versión en español de este libro disponible. Fue traducido por ChatGPT (usando el modelo gpt-3.5-turbo) el 27 de marzo de 2023. Puedes ver la versión en español aquí. Haz clic aquí para ver la solicitud específica que usé para la traducción y la conversación resultante con ChatGPT.

Beginner’s guide to image generation

Ultimately, this guide to image generation should be seen as a stepping stone — a starting point to help you get beyond the blank canvas — aiming to introduce and inspire curious creatives to new tools that can broaden their horizons in the world of graphics and illustrations. It includes a repository of 77 different art styles and sample prompts to try when generating images.

Creating Custom GPTs for journalism

In this guide to creating Custom GPTs, I’ve included dozens of possible use cases for local newsrooms, a step-by-step breakdown of the anatomy and process of building custom GPTs, as well as several examples of custom instructions, best practices, and recommendations for building your own custom GPT assistant.

An updated guide to using AI and automation for local news

This updated guide to using automation and AI for local news emphasizes free and low-cost tools that can be implemented without extensive technical expertise. I’ve included detailed case studies showing how newsrooms are already using AI successfully, from automated newsletter curation to content optimization and administrative task automation.

LLM journalism tool advisor (Replit)

This interactive advisor tool helps journalists select appropriate LLM tools for specific tasks and workflows based on best practices and journalistic requirements. The prototype for the tool was built with Claude 3.7 Sonnet (with extended thinking) and deployed via Replit with help from Replit Agent v2 (beta).

ChatGPT and AI for local news publishers

Amditis presents the first of his AI handbooks for local publishers. During this webinar, he sets the tone for a practical, beginner-friendly journey into generative AI—focusing on how journalists and small newsroom teams can use tools like ChatGPT to improve workflows, reduce burnout, and serve communities more effectively.

Joe introduces ChatGPT, explains the foundational concepts behind large language models, and walks participants through what they can expect from the session. This is a casual, interactive environment aimed at journalists with little to no prior experience using AI tools. The module also touches on ethics, automation, grant-writing use cases, and what it means to integrate AI thoughtfully and responsibly in news production.

How are newsrooms using generative AI for journalism?

A brief overview of how some newsrooms and journalists are using generative AI tools like ChatGPT in their newsgathering and back-office workflows. Produced and narrated by ‪Joe Amditis.

ChatGPT and AI for local news publishers

A workshop that covered the beginner’s guide to custom GPTs from OpenAI, hosted by Joe Amditis. The guide is aimed at small or local newsrooms and organizations looking to improve their resources and staffing.

Amditis provided a detailed overview of GPT and Custom GPT, explaining how they can be used as bots or assistants that can be customized to fit specific needs. He also discussed the challenges of working with AI chatbots, including the recent decline in the quality of responses from OpenAI’s ChatGPT and custom GPTs, and the potential security risks of using third-party apps or programs. Amditis walked through the process of creating a custom GPT assistant using the Create and Configure tabs, using the example of creating a style guide assistant. He provided detailed instructions for customizing the assistant’s behavior and instructions, as well as tips for optimizing its performance.

Amditis also discussed the use of language other than English and the initial configuration instructions, as well as the importance of testing GPTs on different edge cases.

How I created an auto file-sorting automation with Zapier and ChatGPT

I built this prototype to see if it was possible to use Zapier and ChatGPT/Claude 3.5 Sonnet to create an automation that takes any file dropped into a specific Google Drive folder and automatically classifies, describes, tags, and sorts that file into a pre-existing Google Drive folder structure (and create new folders if the appropriate folder doesn’t already exist).

The concept seems valid, though there are definitely several edge cases, specific file types, and other eventualities that I’ll need to consider and work into the process as they arise.

Mastering automation: Zapier for small newsrooms and local journalists (Parts 1-3)

This comprehensive three-part workshop series is designed for journalists, editors, and newsroom managers who are serious about incorporating automation into their daily workflows and newsgathering processes.

Building on the basics, this intermediate Zapier workshop will dive into more complex automation techniques that can significantly boost your newsroom’s efficiency. Learn how to create multi-step Zaps, use filters and conditional logic, and format data within your workflows.

This hands-on session will guide you through building more sophisticated automations tailored to common journalistic tasks. You’ll also learn valuable troubleshooting and optimization tips to ensure your Zaps are running smoothly.

How the Jersey Bee uses AI and automation to produce curated local newsletters

Go behind the scenes of the Jersey Bee’s innovative approach to local news production with founder Simon Galperin. In this detailed walkthrough video, we explore a case study of how a small team leverages AI and automation to serve 12 communities with daily news updates, event listings, and social media content.

Building a basic chatbot with WNYC’s John Keefe

The bots Keefe taught us to build are simple, call-and-response bots that draw on two columns in a Google Sheets spreadsheet in order to receive and respond to user queries. The rest of the back-end infrastructure that informs and guides the behavior of the KeefeBots is still somewhat of a mystery to me, but Keefe did give us a brief introduction to the machine-learning apparatus that operates quietly behind the scenes at wit.ai — I just have a little more work to do before I fully understand it.

What are chatbots and how do they work?

In general, the most effective uses of chatbots seem to be those that are centered around convenience and easy access. Bots that allow users to quickly access the information they desire tend to be the most effective. Therefore, it is important for publications to keep the intent and purpose of their bot in mind when building and implementing it.

What are some ethical considerations that chatbots raise, and how should they be addressed?

Chatbots are an exciting new frontier in machine learning, user engagement, and content distribution. As with any new technology, however, there are ethical challenges and implications that must be considered to ensure that said technology is implemented in a responsible and impactful way.

The bots are coming and you have two options: Adapt or die.

In September 2017, the BBC’s Trushar Barot wrote a lengthy and substantive article for Nieman Lab, in which he argues that voice-activated artificial intelligence represents one of the biggest technological revolutions in the news industry since the first iPhone — possibly bigger. Despite the numerous implications of this emerging technology, Barot writes, publishers have largely ignored this “huge burning platform the news industry doesn’t even know it’s standing on.”

AI throws a lifeline to local publishers (Nieman Journalism Lab)

The ability to quickly generate documents, instructions, guides, and various elements of internal infrastructure could save independent and hyperlocal news publishers an incredible amount of time and effort.

What if you could use AI as the institutional memory for your newsroom?

These kinds of tools aren’t just fancy new toys that can help us do stuff quicker — although they certainly do that, too — they’re also a potential game-changer when it comes to preserving institutional memory. If used properly, I believe they have the potential to revolutionize the way we understand our communities, our body of work, and how we preserve our collective knowledge.

The Nazi chatbots are coming

We’re likely to encounter more people and organizations that are tempted to create fine-tuned models specifically to spread hatred and violence against vulnerable people and marginalized communities. When (not if) that happens, it will likely be disastrous — especially for marginalized and underrepresented communities — further eroding American democratic institutions and exacerbating societal inequities.

AI-generated news has arrived in New Jersey

Can we really rely on machines to produce journalism and news stories that matter? Should we? Can AI really capture the nuances of human life and local color that form the backbone of local news? Is this really the future of local journalism?

Español: Las noticias generadas por IA han llegado a New Jersey

Journalism grapples with the promise and pitfalls of AI-assisted reporting (Nieman Journalism Lab)

As we crawl and claw our way into 2024, we must avoid getting swept away by the potential of these new tools while also anchoring ourselves firmly in the material reality and lived experiences of the communities and people that make our work valuable in the first place. The terrain is littered with ethical landmines and political pressure points that demand more than just an algorithmic touch.

We stop excusing the lack of accessibility in news (Nieman Journalism Lab)

One of the most common excuses for neglecting accessibility is an apparent lack of time or resources. However, advancements in AI tools — I know, I also cringe when someone brings up AI these days, but bear with me for a minute — can actually help get you started.

AI can help journalists but not replace them, experts say

Joe Amditis, assistant director for products and events at the Center for Cooperative Media at Montclair State University in New Jersey, agreed. “I do not recommend that journalists use it to write stories, but it’s very good at helping to edit them,” he said.

Harnessing the power of ChatGPT for local newsrooms

In this episode of Newsroom Robots with Nikita Roy, we explore the potential of ChatGPT for local newsrooms with Joe Amditis, author of Beginner’s Prompt Handbook: ChatGPT for Local News Publishers. Joe is the Assistant Director of Products and Events at the Center for Cooperative Media at Montclair State University, an Adjunct Professor at the School of Communication and Media, and the former producer and host of the WTF Just Happened Today podcast.

The reality of applying AI in newsrooms

In this week’s episode of The Media Copilot podcast, host Pete Pachal explores that question Joe Amditis, Associate Director of Operations at the Center for Cooperative Media. As part of his role, Joe researches how journalists can apply generative AI, both at the individual and organization levels, and has written guides on publicly available tools, including how to create custom GPTs for reporting use cases. Joe’s advice to those new to using AI for journalism? Experiment.

AImpactful #1: Exploring AI in Newsrooms Insights from Joe Amditis (Podcast)

Welcome to the first episode of AImpactful, where we delve into the evolving world of journalism in the age of artificial intelligence. In this enlightening discussion, we are joined by Joe Amditis, an expert in integrating AI into journalistic practices. Joe shares his valuable insights on how journalists can effectively utilize AI tools like ChatGPT to enhance their work. He emphasizes the importance of experimenting with these tools, understanding their capabilities and limitations, and using them to streamline workflow and focus on impactful storytelling. This episode is not just for journalists but for anyone interested in the intersection of technology and media. Join us as we explore the exciting possibilities and challenges AI presents in modern journalism, and gain insights to navigate this new frontier more effectively.

Let AI do the boring stuff so you can focus more on serving your community (Webinar)

Joe Amditis, Center for Cooperative Media at Montclair State University, shows the viewer how to use AI to bridge the apps we use day to day, and take care of the boring work so we can focus on the good stuff.

What concerns does the use of AI in news raise?

As artificial intelligence transforms news production, it offers both innovation and ethical challenges. While AI can streamline content and analyse data, it also raises concerns about misinformation, deepfakes, and accountability. How can the news industry maintain integrity and public trust? This episode explores the potential risks of AI in journalism and the need for robust ethical frameworks to ensure accuracy and transparency in the digital age.

As workplaces adopt AI at varying rates, Gen Z is ahead of the curve

Some experts agree with their assessment that humans aren’t easily replaceable yet. “These [AI] tools should not be put in charge of anything where accountability is required,” said Joe Amditis, associate director of operations for the Center for Cooperative Media at Montclair State University and author of two guides for using AI in local newsrooms. “Because if they can’t pay a fine or if they can’t go to jail or if they can’t be punished or held accountable, then they don’t need to be in charge of decisions that impact people’s lives and livelihoods.”

‘Lean in. Test these things out.’ A lot is being written by and about AI; Here’s some help

In March, CEO Nicholas Thompson wrote a note to his Atlantic business-side staff about AI. “The first point I want to make is that we should all be curious. Lean in. Test these things out. We will be getting some company subscriptions to GPT-4… Secondly, I want us to experiment.” Joe Amditis, who has written a publisher’s ChatGPT guide, said, “If we don’t pay attention to this… we’re going to get tricked ourselves, and we’re going to lose credibility with our audiences.” It’s a lot to navigate right now.