General Education

Some Definitions (from Canadian Association of Journalists' Misinfo 101: National Guide)

Misinformation:

Information that is false or inaccurate, but not created or shared with the intention of causing harm.

• For example: joking with friends about something untrue but funny or unknowingly sharing an article with outdated statistics or the wrong dates with someone.

Disinformation:

Information that is false or inaccurate and is deliberately created or shared to harm a person, social group, organization or country.

• For example: deepfakes.

Malinformation:

Information that is based in reality (but often exaggerated or selectively edited) and shared to inflict harm on a person, social group, organization or country.

• For example: sharing private data or images publicly - "revenge porn"

Image Credit: First Draft News “Understanding Information Disorder,” https://web.archive.org/web/20241009174006/https://firstdraftnews.org/long-form-article/understanding-information-disorder/

Figure 1

Peter Steiner's Famous 1993 New Yorker Cartoon Illustrating an Issue Central to Information Evaluation

Note. From "On the Internet, nobody knows you're a dog" [Cartoon], by P. Steiner, 1993, Wikimedia (https://upload.wikimedia.org/wikipedia/en/f/f8/Internet_dog.jpg).

Evaluating Information

It is good to find lots of search results, but, in order to use information skilfully, you need to know how to evaluate that information to determine whether a specific resource is appropriate to use in a specific use case (i.e. for a specific assignment).

The phrase "evaluating information" actually stands in for a wide range of judgments that we make about information in many different contexts, whether those judgments are about relevance, timeliness, quality, etc.

Librarians have developed several different techniques to help people evaluate information. Two of my personal favourites are RADAR and lateral reading!

RADAR stands for

Relevance

Authority

Date

Accuracy

Reason for Creation

Comparing RADAR with Brian Carroll's "How to Identify Fake News" section (pp. 276-278) from this week's reading:

Step 1: Consider the source. (AUTHORITY)

Step 2: When was the article published? (DATE)

Step 3: Ask about agenda. (REASON FOR CREATION)

- What are the source's sources? (ACCURACY)

Step 4: What is the context?

We can ask the following questions to help us assess each criterion:

Relevance:

• Does this source fit my topic?

• What is this source's intended audience?

  • Is that intended audience appropriate for my use case in this assignment?

Authority:

• Is/are the creator(s) of this source clearly identified or known to us?

• How important is it in this use case to trust the source's creator(s)?

  • If it is important, why should we trust the source's creator(s)?

  • Is the source's creator credentialed or an expert in their field?

Date:

• Is the creation or publication date of this source identified or known to us?

• Is this source too old?

Accuracy:

• Do this source's facts "check out"?

• Does the source have references of its own?

Reason for Creation (take your best guess at this question using judgments from earlier criteria):

• Why was this source made?

• Was this source made to sell a product or service, to inform/educate, to entertain, etc?

(Adapted from Mandalios, J. (2013). RADAR: An approach for helping students evaluate Internet sources. Journal of Information Science, 39(4), 470-478. https://doi.org/10.1177/0165551513478889)

The Five Pillars of Verification (First Draft News)

1. Provenance: Are you looking at the original piece of content?

2. Source: Who captured/created the piece of content?

3. Date: When was the piece of content captured/created?

4. Location: Where was the piece of content captured/created?

5. Motivation: Why was this piece of content captured/created?

Lateral Reading

Lateral reading is an information evaluation technique developed by the Stanford History Education Group (now Digital Inquiry Group) and based on research they conducted comparing the information evaluation habits of Stanford undergraduate students, professors from four different universities, and professional fact checkers.

What the SHEG found in its research was that professional fact checkers spend far less time reading "vertically" - staying on the specific source and evaluating its specific claims. Instead, professional fact checkers would read "laterally" to save themselves time - reading about the first source in other credible sources online to determine what the first source's potential biases may be, how credible it is, etc.

Generative Artificial Intelligence, Misinformation, and Disinformation

Generative artificial intelligence (AI) is a hot topic these days that is having an impact on many areas of cultural life, education, and the economy.

Generative AI tools can be used to create and spread misinformation and disinformation in multiple ways.

In NewsGuard's October 2024 audit of the 10 leading GenAI chatbots, the organization found that those chatbots "collectively repeated misinformation 24.67 percent of the time, offered a non-response 21.67 percent of the time, and a debunk 53.67 percent of the time. The 46.33 percent “fail” rate (percentage of responses containing misinformation or offering a non-response) increased from NewsGuard’s September audit."

In addition, as of November 11, 2024, NewsGuard's team has identified 1,121 unreliable AI-generated news and information websites spanning 16 languages.

OpenAI, the company behind ChatGPT, states, in its own documentation for GPT-4, that “[t]he profusion of false information from [large language models] … has the potential to cast doubt on the whole information environment, threatening our ability to distinguish fact from fiction.”

Two ways that this has happened already and that have made the news have been through fake image generation and textual fabrication.

A Specific Textual Fabrication Example From ChatGPT

Be sure to scrutinize any source(s) that generative AI provides you with on a topic. This is because, at this point, it is prone to error: what some have called "hallucination," but that I prefer to call "fabrication."

If generative AI provides you with a source:

(1) make sure that the source actually exists, and, if it does exist;

(2) make sure that the source actually contains the information that generative AI has attributed to it.

An Investigation of ChatGPT's Sources

1. Book: Influencer Marketing for Dummies by Kristy Sammis, Cat Lincoln, and Stefania Pomponi

   • This source does exist and it was written by these authors, but it is a For Dummies book that wouldn't be considered scholarly.

2. Book: Influencer Marketing: Building Brand in a Digital Age by Duncan Brown and Nick Haye

   • This source does exist and it was written by those authors, but ChatGPT has fabricated a subtitle for it that it doesn't have.

3. Academic Article: "The Rise of Influencer Marketing and Its Impact on Consumer Behavior" by Liu, Hu, and Zhang (2019)

   • To the best of my searching abilities, this source does not exist.

4. Academic Article: "Ethical and Legal Issues in Influencer Marketing" by Brenner, A. and Capron, L. (2019)

   • To the best of my searching abilities, this source does not exist.

5. Academic Article: "The Dark Side of Social Media: A Consumer Psychology Perspective" by Phua, J., Jin, S.V., and Kim, J.J. (2017)

   • This source is a Frankenstein composite of 2 sources. The authors have been taken from this article and the title has been taken from this edited book with which those authors had no involvement.

GenAI & Mis/Disinformation Activity

Here is a Padlet board to use to track your research: https://padlet.com/bleching/gned1403_genai

1. Please put your names below an item in a comment to claim your information source.

2. Next, perform a "lateral reading" of your source by searching the web for it and finding other reputable sources that talk about it or that are hosting it online.

Useful websites that you might use to analyze your source might be:

• Tineye

• Google's Reverse Image Search (instructions here) or (in Chrome) right click and "Search with Google Lens."

• You might also want to use Hive Moderation's AI-Generated Content Detector.

If your source has text, you may want to also search that text on Google in quotes to find more info about it. For example:

"example text"

3. Write what you find in other comments and try to answer some of these questions:

• Who created your source (if applicable/known)?

• Why did they create it (if applicable/known)?

• If you find that your source was AI generated, which GenAI tool(s) were used to create it (if applicable/known)?

• What issues related to GenAI does your source represent?

4. We will wrap up in ~10-12 mins.

• Source 1

• Source 2

• Source 3

• Source 4

• Source 5

• Source 6

• Source 7

• Source 8

• Source 9

• Source 10

• Source 11

TikTok and Misinformation/Disinformation

Given the video sharing platform TikTok's widespread popularity right now, it is full of misinformation and disinformation that can travel at lightning speed as bite-sized clips.

NewsGuard, an organization that tracks mis/disinformation online, performed an analysis of TikTok content in 2022. The organization found that the platform “repeatedly delivered videos containing false claims in the first 20 results, often within the first five" to users. Out of 540 TikTok videos that NewsGuard vetted for its report, 105 (19.4%) contained false or misleading claims.

Though the company claims to be combatting false information through collaboration with fact-checking organizations like PolitiFact and Reuters, there is still tremendous potential for misinformation and disinformation to spread on TikTok due to the nature of virality and how much content is circulating on the platform.

General TikTok Information Literacy Principles

- Be aware of the potential for the selective editing of video and video manipulation.

- Be aware of the potential for the selective editing of audio and audio manipulation.

- Consider how misinformation and disinformation from other platforms can easily spread on TikTok. (Example).

Further Resources

• NewsGuard Reports

• Civic Online Reasoning (Digital Inquiry Group)

• Media Smarts (Canada's Centre for Digital Media Literacy)

• Doubt It? (NewsWise)

• Spot Fake News (News Media Canada)

• Canadian Association of Journalists' Misinfo 101 National Guide

• Interactive Media Bias Chart (Ad Fontes Media)

• Deepfakes Tracker Research Portal (Social Media Lab at Toronto Metropolitan University)

• MRUnderstanding Misinformation (Developed by Dr. Brooks DeCillia - Journalism faculty here at MRU)