AI Explorations

As AI becomes more inte­grat­ed into the post-sec­ondary expe­ri­ence, it’s impor­tant that instruc­tors acquire knowl­edge of the capa­bil­i­ties of AI and skill in the avail­able tools that could be uti­lized by stu­dents. Focus­ing on explor­ing, play­ing, shar­ing, ask­ing ques­tions, and learn­ing about arti­fi­cial intel­li­gence and its impact on post-sec­ond edu­ca­tion is key as you build a deep­er under­stand­ing of this top­ic.

The fol­low­ing ‘Explo­rations’ will focus on teach­ing and learn­ing com­po­nents and chal­lenge you with tasks to com­plete to become more famil­iar with spe­cif­ic tools. The last 5 explo­rations align with one of 6 AI laneways that are iden­ti­fied in the NIC AI Guide­lines.

Video Expla­na­tion (6 mins)

Ethan Mol­lick­’s book, “Co-Intel­li­gence: Liv­ing and Work­ing with AI” was pub­lished in April 2024 and has been respect­ed as an excel­lent starter book in under­stand­ing AI. It was the cho­sen book back in May for Cana­di­an high­er ed edu­ca­tors and teach­ing lead­ers to read and explore via a pan-Cana­di­an explo­ration group. The NIC library has hard copies in all of the cam­pus­es for sign­ing out. This 4–5 hour breezy read is high­ly rec­om­mend­ed. It isn’t edu­ca­tion spe­cif­ic but has sep­a­rate chap­ters on var­i­ous poten­tial uses of AI with exam­ples he shares. He writes in very acces­si­ble lan­guage.

Ethan is often con­sid­ered a bal­anced, but opti­mistic post-sec­ondary edu­ca­tor who (along with his wife) are at the fore­front of explor­ing how AI can impact edu­ca­tion, work and life — along with the pit­falls, pri­va­cy, secu­ri­ty and oth­er uncer­tain com­po­nents still being uncov­ered. He does a lot of dai­ly work behind the scenes in explor­ing AI. He is a busi­ness pro­fes­sor at Uni­ver­si­ty of Penn­syl­va­nia. He writes a blog post on Sub­stack, has a cou­ple of web­sites shar­ing his work and often posts on LinkedIn.

His two web­sites: One Use­ful Thing and More Use­ful Things are excel­lent sites to read his lat­est blog post (One Use­ful Thing) or check out his Prompt Library and oth­er resources (More Use­ful Things). You will be asked to sub­scribe to his Sub­stack or upgrade to a paid ver­sion- you can agree to freely receive his posts in your email or ignore pop up box.

To kick start your AI Explo­ration read Ethan’s lat­est (100th) post on his Sub­stack called, “Think­ing Like an AI”- A lit­tle intu­ition can help (Octo­ber 20, 2024). In sim­ple lan­guage he describes how AI works. Length: 2500 words — may take you any­where from 6 min­utes to 16 min­utes to read.

After you have read the arti­cle, try out one or two of his pre-writ­ten prompts: for Instruc­tors or for Stu­dents. He indi­cates which of the most com­mon GPTs (Gen­er­a­tive Pre-trained Trans­form­ers) work best with each prompt (Chat­G­PT, Bing/Copilot, Gem­i­ni or Claude).

Ethan has essen­tial­ly cre­at­ed ‘text’ for you to just copy and paste into a GenAI plat­form with all the key terms to have it turned into an assis­tant, a tutor, a quiz cre­ator, a sim­u­la­tion cre­ator, a les­son crafter etc. This is often where peo­ple strug­gle the most — writ­ing good instruc­tions for GPT mod­els. There are full length cours­es just on how to ‘write a prompt’ but Ethan gives you quite a num­ber that you can use and they work!

No need to pur­chase any­thing — free accounts will do for most of his prompts (paid ver­sions have more fea­tures and capac­i­ty but let’s leave that for anoth­er day — they are often labelled as “Advanced”).

If you don’t wish to cre­ate a free account use Microsoft Copi­lot which you have with your NIC account or Chat­G­PT. Don’t put any per­son­al infor­ma­tion, stu­dent work, con­fi­den­tial infor­ma­tion into GPTs!

But: some­time it might be good to get your­self a set of free accounts of the three oth­er GPTs (chat envi­ron­ments) for our explo­rations — as each works dif­fer­ent­ly and it might be good learn­ing to see what one can and can’t do.

  • Chat­G­PT or GPT4 (from com­pa­ny Ope­nAI — free no account ver­sion, or look to top right and cre­ate a free account)
  • Claude (from com­pa­ny Anthrop­ic — free with account you cre­ate with email)
  • Gem­i­ni (from Google, used to be called Bard — free ver­sion with your Gmail/Google account when you sign in).
  • Copi­lot (from com­pa­ny Microsoft, used to be called Bing Chat — free ver­sion with your NIC account when you are signed in)

So for this explo­ration let’s try this and fol­low these pro­to­cols:

  • Take 6–16 min­utes to read: Think­ing Like an AI
  • Explore some of the pre-writ­ten prompts (Instruc­tor and Stu­dent ones) in Copi­lot or oth­er GPTs from More Use­ful Things: Prompt Library - by copy­ing the prompt into the first chat box, allow­ing it to think and then fol­low its respons­es as if you were using it as an instruc­tor or stu­dent (this takes min­utes to do but you may wish to fol­low the prompts and try out in a vari­ety of GPTs or dif­fer­ent prompts to see dif­fer­ences)

Many moons ago as a uni­ver­si­ty pro­fes­sor, Liesel Knaack did research on “learn­ing objects” (assess­ment of them, how to use them in learn­ing process etc.) and taught an advanced instruc­tion­al design elec­tive build­ing learn­ing objects. Learn­ing objects are inter­ac­tive edu­ca­tion­al tools or col­lec­tion of content/practice/assessment items that are freely acces­si­ble to any­one to share and reuse in many cours­es.

Her stu­dents would use Adobe Flash (a for­mer mul­ti­me­dia plat­form) to build these learn­ing objects/resources that they couldn’t find online or that the text­book or sta­t­ic mate­ri­als pre­vent­ed learn­ers from ful­ly under­stand­ing a con­cept. It took them nine weeks of class time and home­work to build a learn­ing object that func­tioned prop­er­ly. Con­cepts such as bal­anc­ing equa­tions in chem­istry or under­stand­ing how a dan­ger­ous sim­u­la­tion might unfold, how to pre­vent pla­gia­rism or a physics law that required stu­dents to input var­i­ous val­ues to see the out­come.

Her stu­dents did such a great job that she took them to the MERLOT (mul­ti­me­dia shar­ing plat­form) con­fer­ence in Ottawa one year and they pre­sent­ed their prod­ucts — see some of their exam­ples with­in slide deck. This was the begin­ning of her pas­sion for cre­at­ing share­able, reusable, remix­able learn­ing tools that even­tu­al­ly would expand into Open Edu­ca­tion­al Resources (OERs). 

Skip for­ward two decades and we see learn­ing object exam­ples such as sim­u­la­tions devel­oped by the Uni­ver­si­ty of Col­orado to help stu­dents under­stand physics — or stu­dents explore math cal­cu­la­tions or engage in sim­u­la­tions and vir­tu­al labs.

While the web now has many of these learn­ing objects avail­able, there are small­er inter­ac­tive com­po­nents that teach­ers look to include in their cours­es to help stu­dents learn or they just want some­thing they can’t find online (a tool or a demo). Or teach­ers want some small­er com­po­nents auto­mat­ed or cre­at­ed for engage­ment and prac­tice by stu­dents.

Her inter­est in these inter­ac­tive com­po­nents remained evi­dent as she shared how to use H5P (HTML 5 Pack­age) in build­ing inter­ac­tive activ­i­ties (quiz ques­tions, match­ing games, con­tent design, engag­ing pre­sen­ta­tions etc.) through var­i­ous work­shops held in 2024 at NIC. 

But recent­ly Claude has the abil­i­ty to cre­ate “arti­facts”. An arti­fact can be more than a H5P inter­ac­tive com­po­nent — it could be doc­u­ment, snip­pets of code, a whole web­page (HTML code) or it can be an image or dia­gram – but what Liesel liked were the inter­ac­tive “React” com­po­nents like learn­ing objects.

Here is a sim­ple one she made to cal­cu­late read­ing time based on words. Kind of like a ‘cal­cu­la­tor’ where it takes the research on read­ing time (which she con­firmed is cor­rect) and makes a lit­tle input device to approx­i­mate read­ing time for assign­ments or read­ings.

Here is anoth­er real­ly quick (but kin­da hard) inter­ac­tive Claude Arti­fact she made – called a Gen­er­a­tive AI Match­ing Game – you click on two cards and try to find the def­i­n­i­tion and term and match them. When you get them cor­rect they stay turned over. You do the game until you match all terms and def­i­n­i­tions. 

But how did she get there?

She asked (free) Claude with this prompt:
Hi Claude. Can you cre­ate a web tool that acts like a match­ing game of cards. There are 16 cards with 8 pairs that match. The top­ic is gen­er­a­tive AI def­i­n­i­tions. One of the cards in a pair is the word and anoth­er card is the def­i­n­i­tion. The play­er has to click and turn over only two cards at once and if they are cor­rect pair (def­i­n­i­tion and word) and the play­er gets a point. If the cards are not a pair then they are turned back over in the same spot and the next play­er tries.

Claude respond­ed with:
I’ll cre­ate an inter­ac­tive card match­ing game focused on gen­er­a­tive AI terms using React. This will include card flip­ping ani­ma­tions, score track­ing, and the match­ing log­ic you described. And then in the win­dow to the right it start­ed doing all the cod­ing. A minute lat­er it built the match­ing game. I did noth­ing but watch!

Then she got this from Claude:
I’ve cre­at­ed a React com­po­nent that imple­ments the Gen­er­a­tive AI Match­ing Game you described. Here’s a break­down of the game fea­tures:

  1. The game con­sists of 16 cards (8 pairs) with gen­er­a­tive AI terms and their def­i­n­i­tions.
  2. Cards are shuf­fled at the start of the game and dis­played in a 4x4 grid.
  3. Play­ers can flip two cards at a time by click­ing on them.
  4. If the flipped cards match (a term and its def­i­n­i­tion), they remain face up, and the play­er scores a point.
  5. If the cards don’t match, they flip back after a short delay.
  6. The game tracks the play­er’s score.
  7. There’s a “Shuf­fle and Restart” but­ton to reset the game at any time.
  8. When all pairs are matched, a con­grat­u­la­to­ry mes­sage appears.

The com­po­nent uses some UI com­po­nents from the shadcn/ui library (But­ton and Card) for styling. It also uses the Shuf­fle icon from the lucide-react library for the reset but­ton. To use this com­po­nent, you’ll need to ensure that the nec­es­sary depen­den­cies are installed and that you have the shadcn/ui com­po­nents set up in your project. Would you like me to explain any part of the code in more detail or make any mod­i­fi­ca­tions to the game?

She saw the game, played with it a few times and then saw the abil­i­ty to pub­lish it – upon pub­lish­ing she got a link to share with oth­ers. She could put it into her course or share it on a web­site.

It isn’t fan­cy – but it is a learn­ing object she built with a sim­ple prompt and no cod­ing – in min­utes! Remem­ber: there could be errors or hal­lu­ci­na­tions!

TRY build­ing an arti­fact in Claude — some­thing more adven­ture­some than a match­ing card game please! Pub­lish and share your link to your arti­fact so we can try it out. Share how you might find it help­ful. Share what you learned? Here are some ideas.

NOTE: Free ver­sion of Claude restricts you to 7 prompts (I think) until it won’t do any more until you pur­chase next ver­sion. So think of your prompts or start a new con­ver­sa­tion if you run out of “tokens” in a giv­en con­ver­sa­tion.

Resources

Google Note­bookLM is an AI tool from Google (that was released over a year ago) — that can ana­lyze up to 50 sources (web­site links, PDFs, YouTube videos) and extract themes, key mes­sages and impor­tant points. It does­n’t do research on the web — it focus­es sole­ly on the sources you upload. It is an AI note­book — a handy dandy research assis­tant!

Just this past month a new fea­ture was added to Google Note­bookLM and puts this in a whole new cat­e­go­ry of research sup­port. 

The biggest fea­ture (that caused jaw drop­ping respons­es when it came out) is the cre­ation of a PODCAST (with two real­is­tic voic­es) chat­ting about the con­tent you have uploaded.  Seri­ous­ly. You can take excit­ing, dry or dense con­tent and upload it and the pod­cast will gen­er­ate (it takes a few mins to cre­ate pod­cast) a live­ly dis­cus­sion about that con­tent. You got­ta try it! Time Mag­a­zine called it one of the best inven­tions of 2024. 

Oth­er fea­tures include the abil­i­ty to com­bine images, PDFs, and oth­er con­tent into instant research papers, cur­ricu­lum for a course, etc. It also takes con­tent in a vari­ety of for­mats — and it ref­er­ences where the details came from along with table of con­tents, brief­ing doc­u­ments and study guides! That’s the research assis­tant part — so if you can get past the pod­cast fea­ture there is much more here to explore. 

Explore how it can build a research paper, orga­nize it and build a final prod­uct. It is a time sav­ing tool for busy aca­d­e­mics!

Try it out: https://notebooklm.google/ (you need a Google account)

Rec­om­mend­ed Skims

Your turn! Try out Google Note­bookLM — and from an edu­ca­tion­al per­spec­tive of help­ing you with orga­niz­ing papers, cre­at­ing out­lines, them­ing con­tent from any source you upload (and the pod­cast fea­ture) — HOW could stu­dents use this in an assign­ment (as an allow­able tool to use) and how could YOU (as an edu­ca­tor, employ­ee) use this for your work in prepar­ing teach­ing and learn­ing activ­i­ties?

Well, we can’t go much fur­ther in our AI explo­rations with­out giv­ing some air­time to the cre­ative aspect of image gen­er­a­tion (text to image) – but with a healthy dose of com­ing up against deep­fakes!

This week take the deep­fake quiz (to see if you can spot the real pic­tures!) and then go out explor­ing and cre­ate some real­is­tic and/or fan­tas­ti­cal images to use in your course.

Remem­ber: All free accounts will count “tokens” (pro­cess­ing pow­er to build images) and after a cou­ple of pic­tures may cut you off until 24 hours have passed. So, think about your prompt before you go muck­ing about. You can get a few decent images on all free accounts with the plat­forms below.

FIRST — TAKE QUIZ:

Take the deep­fake 10 ques­tion quiz to see if you can spot the deep­fakes (real look­ing pic­tures that were com­put­er gen­er­at­ed) at Leon Furze’s site.

SECOND — SKIM ARTICLE:

This arti­cle (with accor­dion expand­ing show­ing same images test­ed in each plat­form) arti­cle from ZDNET on their reviews of image gen­er­a­tion GPTs — with­in each review of the plat­form there is a link to anoth­er arti­cle with steps on how to use the plat­form.

THIRD — EXPLORE:

AI Image Gen­er­a­tors (some of them)

Open AI’s Dall‑E 3 — Web­site

  • CNET Review Page
  • Cost: $20 a month with Chat­G­PT Plus Sub­scrip­tion
  • Chat­G­PT Free Ver­sion: 2 images in a 24-hour peri­od

Google’s Image­FX – Web­site

  • Through Google Account
  • ZDNET Review and Instruc­tions to Use
  • Can access this through Gem­i­ni chat­bot

Leonar­do AI – Web­site

  • CNET Review of Leonar­do AI
  • Cost: Free and pre­mi­um plans
  • 150 tokens that resets dai­ly
  • Options for size and num­ber of images per each gen­er­a­tion

Microsoft Designer’s AI Image Gen­er­a­tor — Web­site

  • Cost: Free ver­sion with NIC Microsoft license
  • ZDNET Review and instruc­tions

Adobe Fire­fly – Web­site

Mid­jour­ney — Web­site

  • No free account – starts at 10$ US/month
  • Very high-qual­i­ty images

Can­va — Web­site

  • Cost: Free and paid account
  • Good basic image gen­er­ate for begin­ners

Video and audio is an untapped poten­tial in post-sec­ondary edu­ca­tion but has been chal­leng­ing to access due to the learn­ing curve and the need for sophis­ti­cat­ed soft­ware and hard­ware resources required to engage stu­dents with the plat­forms and their out­puts. NO MORE!

Liesel Knaack was so inspired by these plat­forms that she took a year off work (eons ago) to com­plete a post-grad­u­ate cer­tifi­cate in mul­ti-media at col­lege. She got an immer­sive learn­ing expe­ri­ence shoot­ing video with pro­duc­tion lev­el cam­eras, doing audio pro­duc­tion in a stu­dio and cre­at­ing a cou­ple of inter­ac­tive edu­ca­tion­al resources for learn­ers. 

In her ear­ly days of post-sec­ondary teach­ing she had her stu­dents mak­ing videos, cre­at­ing pod­casts and devel­op­ing inter­ac­tive assign­ments — always full of cre­ativ­i­ty and fun in learn­ing. In mid 2000’s she got into Jing (now called Tech­Smith Cap­ture) — a light­weight (and free) video record­ing tool (like Kaltura Cap­ture which you have access in Bright­space and NIC Medi­a­Space)  to record sum­maries for her class­es, cre­ate instruc­tion­al videos and bring con­tent to life. Here are oth­er uses of video.

She sees many uses of text to video gen­er­a­tion for instruc­tors and espe­cial­ly for those who cre­ate train­ing mate­ri­als, pre­sen­ta­tions, intro­duc­to­ry videos — as you can cre­ate stu­dio-qual­i­ty videos with AI avatars and voiceovers in many lan­guages — and as easy as Pow­er­Point. If you don’t have a video-qual­i­ty voice or image or don’t speak many lan­guages or don’t want to waste hours cut­ting videos — then check this out!

Here is what Liesel made in 20 min­utes with Syn­the­sia: MOST of the text was pro­vid­ed but she zhuzhed it up a bit! (or accessed from host)

She set up a free account with Syn­the­sia (eas­i­er to use than oth­er pop­u­lar prod­uct, Hey­Gen). Upon account set up, she shared what she was want­i­ng to do (cre­ate train­ing videos, pro­mote a top­ic). 

Then for the project top­ic it asked her what she want­ed to cre­ate. She said “cre­ate a video out­lin­ing the pos­si­bil­i­ties with gen­er­a­tive AI text to video for teach­ing and learn­ing”. 

With­in sec­onds it cre­at­ed the slides, the text, the images, the builds, the tran­si­tions and the default avatar. She changed up the text, added anoth­er slide, repo­si­tioned the avatar, pre­viewed what each nar­ra­tion would sound like, took off the logo and gen­er­at­ed the 2.5 minute video in 9 min­utes. She down­loaded and uploaded to Kaltura Video. You don’t need to do that — you can just “pub­lish” it and it will store on their site with a URL.

With paid accounts, you could use over 180 avatars, have up to 360 min­utes of video per year — and cre­ate your own dig­i­tal twin (your own avatar of you!).The free account is only 3 min­utes in total per month with 9 AI avatar choic­es and 2 stock per­son­al avatars — but you can access the 140 lan­guages and voic­es!

Overview for train­ing and learnin — Give it a try!

If you are not into the video aspect but want to try text to audio — Eleven Labs is a great start­ing spot with a free account to cre­ate some­thing like this short audio clip or this sound effect Liesel made with­in min­utes! Eleven Labs is a voice gen­er­a­tor that can pro­duce con­tent in 32 lan­guages — great for any voice overs, cre­at­ing an audio book or pro­duc­ing con­tent for a course that is in a for­mat that stu­dents can lis­ten to — like a mod­ule audio book!

Explore
1. Cre­ate a video from text with avatar: Use Syn­the­sia (or Hey­Gen or oth­er plat­forms) and/or
2. Cre­ate an audio clip from text: Use Eleven Labs or any oth­er sug­gest­ed plat­forms like Speechi­fy, Altered and Murf
3. Reflect on how you might use the end prod­uct in teach­ing and learn­ing sit­u­a­tions

As the enor­mi­ty of AI inte­gra­tion in our every­day lives becomes more promi­nent and acces­si­ble we need to con­sid­er how it can be tai­lored to ben­e­fit our stu­dents as they progress through high­er edu­ca­tion stud­ies. Through the adop­tion of a 6‑Lane AI Assess­ment Scale here at NIC we can see many areas where strate­gic, inten­tion­al inte­gra­tion of AI tools can assist learn­ers.

This week’s post and chal­lenge will focus on Lane 2: AI as a Study Tool. There are many ways that AI can be used as a study tool, specif­i­cal­ly this week we are going to explore the prac­tice of tutor­ing.

Who has not ben­e­fit­ed from tutor­ing in their own per­son­al learn­ing paths? Many break­through moments of under­stand­ing are in deep con­ver­sa­tion with experts on a par­tic­u­lar top­ic.

How often do your stu­dents have these oppor­tu­ni­ties? It’s safe to guess that an expert is not read­i­ly avail­able to inter­act with each of your stu­dents to pro­vide unend­ing atten­tion and patience in coach­ing them along their path of under­stand­ing. Most like­ly, you end up fill­ing this role. An unat­tain­able goal this is for an instruc­tor to indi­vid­u­al­ly tutor each one of their stu­dents while also teach­ing the class. The next log­i­cal place to look for this assis­tance was beyond our class­rooms to peer tutor cen­ters and some­times pri­vate­ly paid tutors. Up until recent­ly, a phys­i­cal tutor was the only option avail­able along with a mine­field of lim­it­ed tutor num­bers avail­able, lim­it­ed exper­tise, restrict­ed time slots and (some­times) high cost. How are we as edu­ca­tors to pro­vide every stu­dent the same oppor­tu­ni­ties as Ben­jamin Bloom iden­ti­fied back in 1984 when he doc­u­ment­ed that tutored stu­dents achieved two grade point aver­ages high­er than their un-tutored peers? Enter AI tutor­ing.

AI can assist stu­dents in the role of a tutor at any time of the day, any­where geo­graph­i­cal­ly for zero cost. I agree, it is not the same as a human per­son­al tutor. But hon­est­ly, isn’t this step bet­ter than no tutor at all?

Below are 4 different options to get you and your students moving with AI as a tutoring study tool:

#1 Cre­ate your own tutor­ing prompt. The AI must under­stand that it is an expert in the top­ic and explic­it­ly describe tutor­ing. One of the exhaust­ing parts of mak­ing the AI do this task is to frame the expec­ta­tion of the tutor­ing expe­ri­ence as full as pos­si­ble. The longer the prompt, the bet­ter. Try using some of these:

  • Ask me ques­tions
  • Do not imme­di­ate­ly pro­vide an answer
  • If I respond incor­rect­ly ask me why I think that is the answer
  • Do not pro­vide the entire process in one response. Break it down and only pro­vide one part at a time after I have had a chance to respond
  • Explain tricky con­cepts in two dif­fer­ent ways
  • Use analo­gies or metaphors in the expla­na­tions
  • Gauge my base­line under­stand­ing by ask­ing 5 ques­tions pri­or to start­ing the tutor­ing

#2 Pre-craft­ed Tutor­ing prompt. Why devel­op your own if some­one has already done it for you? For instance, the most recent AI tutor prompt from Ethan Mol­lick is an aston­ish­ing 791 words in length! Yeow. Ethan’s web­site More Use­ful Things: AI Resources con­tains a handy prompt library with the afore­men­tioned behe­moth of a prompt and many, many more equal in detail. This is well worth the time to sim­ply explore.

#3 Real-time audio/visual AI tutor­ing. AI tutor­ing through text may be an excel­lent for­mat for top­ics that can be eas­i­ly expressed in writ­ten form, but what about physics, math­e­mat­ics, or dia­grams? Per­haps the stu­dent has spent 2hrs on a process or cal­cu­la­tion and can­not under­stand where they are going wrong. What would a tutor typ­i­cal­ly do in this sit­u­a­tion?  Per­haps they would start by review­ing the stu­den­t’s work.

The newest release of Gem­i­ni can access the micro­phone & web­cam of a com­put­er and ana­lyze what it sees and hears in real-time. If the appli­ca­tion is used on a smart­phone the func­tion­al­i­ty is rem­i­nis­cent of shar­ing one’s sur­round­ings with a friend on Face­time.  Stream Real­time | Google AI Stu­dio

In addi­tion, Gem­i­ni now has built-in tutors for spe­cif­ic top­ics such as math­e­mat­ics all ready to go.

#4 Free AI Tutor Plat­form. Here is yet anoth­er option, use AI Tutor Pro from Ontario-based Con­tact Nord. No need to pro­vide detailed prompts, it has already been designed for tutor­ing. Stu­dents can choose to type in a sub­ject or upload con­tent about a con­cept or top­ic they are learn­ing. Free.

AI Tutor Options for $$$:

Chat GPT has released an edu­ca­tion-spe­cif­ic ver­sion of its GPT-4o plat­form that can be pur­chased by post-sec­ondary edu­ca­tion providers. Per­haps this is the direc­tion high­er edu­ca­tion will go. See the site here: Ope­nAI for Edu­ca­tion | Ope­nAI.

Khan Acad­e­my has released a tutor­ing-spe­cif­ic appli­ca­tion for any­one. Some school dis­tricts in the USA are pilot­ing its use in assist­ing stu­dent. This cur­rent­ly requires a fee to access in Cana­da. See the site here: Khan­mi­go.

How do you see these resources being used in your course? How can this resource help empow­er your stu­dents to learn more deeply? Try it out for your­self…

  1. Choose one method out­lined above to inter­act with an AI tutor.
  2. Choose a top­ic
  3. Expe­ri­ence the AI tutor process

Resources:

• From Khan Acad­e­my: Meet Khan­mi­go: Khan Acad­e­my’s AI-pow­ered teach­ing assis­tant & tutor
• Blog Post: How AI Could Save (Not Destroy) Edu­ca­tion | Sal Khan | TED
• TeachAny­where: Prompts, Tutors and Assis­tants

This explo­ration focus­es on Lane 4: AI as edi­tor. AI tools can help stu­dents improve their writ­ing by check­ing gram­mar, sug­gest­ing bet­ter word choic­es, and mak­ing their ideas clear­er. Edit­ing tools also help stu­dents catch mis­takes and fix them before turn­ing in their work.

Through­out post-sec­ondary edu­ca­tion expe­ri­ences, many of us have ben­e­fit­ted from col­leagues, peers, and pro­fes­sors who pro­vid­ed feed­back on our work. Liesel Knaack still attrib­ut­es her first break­through as a writer dur­ing her under­grad­u­ate when her cul­tur­al geog­ra­phy pro­fes­sor pro­vid­ed edit­ing sug­ges­tions and numer­ous oppor­tu­ni­ties to redo her work. Sim­i­lar­ly, when she jour­neyed through post-grad pro­grams, she had advi­sors and com­mit­tees who took the time to pro­vide feed­back on her writ­ing projects.

Her expe­ri­ences are now 10 + years ago and times have changed. For one, we lived through a pan­dem­ic that sig­nif­i­cant­ly impact­ed how we study and learn togeth­er. Changes such as a shift to remote learn­ing and using dig­i­tal tools accel­er­at­ed the arrival of AI edit­ing tools. AI as an edi­tor dove­tails with AI as a tutor as both enhance stu­dents’ capa­bil­i­ties to refine writ­ten con­tent and facil­i­tate learn­ing. AI tutors and AI edit­ing tools have made learn­ing and con­tent cre­ation more acces­si­ble and con­ve­nient for those with access to and famil­iar­i­ty with the tech­nol­o­gy.

So how do we use AI as an edit­ing tool? 

Sim­ple. 

AI edit­ing tools can help refine and per­fect writ­ten con­tent through gram­mar and spell check. Tools can check for sen­tence struc­ture and clar­i­ty, style, and tone. Some AI edit­ing tools offer con­tent sug­ges­tions, such as rec­om­mend­ing alter­na­tive phras­ing, sen­tences, and even entire para­graphs. Bot­tom line, AI edit­ing tools assist by flag­ging poten­tial issues, pro­vid­ing sug­ges­tions, and stream­lin­ing the edit­ing process. 

How can instruc­tors use AI with stu­dents’ assign­ments?

AI edit­ing tools pro­vide an oppor­tu­ni­ty for instruc­tors to guide stu­dents not only in refin­ing their writ­ing but also in devel­op­ing crit­i­cal think­ing about writ­ing qual­i­ty. Here are some ways instruc­tors can incor­po­rate AI edit­ing tools into assign­ments:

  • Pre­lim­i­nary Edits: Assign stu­dents to run their drafts through AI tools like Microsoft Copi­lot, Gram­marly, or Chat­G­PT before sub­mis­sion. Ask them to sub­mit both the orig­i­nal and the AI-edit­ed ver­sion, along with a reflec­tion on how the edits improved their work or aligned with the assign­men­t’s goals.
  • Revi­sion Work­shops: Encour­age peer review using AI. Stu­dents can share their drafts and use AI-gen­er­at­ed feed­back as a sup­ple­ment to their peers’ sug­ges­tions, pro­mot­ing col­lab­o­ra­tive and tech­no­log­i­cal­ly enhanced learn­ing.
  • Tone and Style Exer­cis­es: Assign stu­dents a piece of their writ­ing to ana­lyze using AI. For instance, ask them to use Notion AI to shift the tone from for­mal to con­ver­sa­tion­al, or refine clar­i­ty and con­cise­ness, then dis­cuss how those changes impact their mes­sage.
  • Gram­mar and Struc­ture Analy­sis: Cre­ate assign­ments where stu­dents com­pare AI edit­ing sug­ges­tions with tra­di­tion­al gram­mar guides. This can help them crit­i­cal­ly eval­u­ate the reli­a­bil­i­ty and appro­pri­ate­ness of AI feed­back.
  • AI Lim­i­ta­tions Dis­cus­sions: Include assign­ments where stu­dents ana­lyze the lim­i­ta­tions of AI tools. For exam­ple, they can iden­ti­fy areas where the tool mis­un­der­stood con­text or offered unsuit­able sug­ges­tions.

By fram­ing assign­ments around these tools, instruc­tors can teach stu­dents not only how to use AI effec­tive­ly but also to approach these tools crit­i­cal­ly, ensur­ing they remain active par­tic­i­pants in their learn­ing process.

Below are 4 dif­fer­ent options to get you and your stu­dents to use AI as an edit­ing tool.

Microsoft Copi­lot is like­ly one of the eas­i­est edit­ing tools for stu­dents to use, espe­cial­ly if they are already famil­iar with Microsoft Office appli­ca­tions like Word. Famil­iar­i­ty means stu­dents don’t need to learn a new tool, they can access Copi­lot direct­ly with­in the soft­ware they use for assign­ments.

Here are step-by-step instruc­tions for using Microsoft Copi­lot as an edit­ing tool in Word or oth­er Office appli­ca­tions:

Step 1: Open the Doc­u­ment

  • Open the doc­u­ment you want to edit, select the text and copy into copi­lot.

Step 2: Select the Text to Edit

  • High­light sec­tions you want help with or place your cur­sor where you need edit­ing assis­tance.
  • For doc­u­ment-wide edits, you don’t need to high­light any­thing, Copi­lot can ana­lyze the whole text.

Step 3: Choose an Edit­ing Option

  • Select from the avail­able fea­tures, such as:
  • Gram­mar and Spelling Check
  • Clar­i­ty and Tone Sug­ges­tions
  • Rewrit­ing Sug­ges­tions:
  • Apply these sug­ges­tions indi­vid­u­al­ly or use bulk approval options.

Step 4: Review and Apply Sug­ges­tions

  • Review each sug­ges­tion Copi­lot pro­vides.
  • Accept, reject, or mod­i­fy sug­ges­tions based on your pref­er­ences.

Step 5: Final­ize and Save

  • Once you’ve made all edits, proof­read your work to ensure it meets your expec­ta­tions.
  • Save the doc­u­ment with a new name if you want to keep the orig­i­nal ver­sion as well.

Prac­tice Using Copi­lot: The more you use it, the more famil­iar and effi­cient you’ll become with its capa­bil­i­ties!

Chat­G­PT can be an effec­tive edit­ing tool for stu­dents, offer­ing assis­tance with gram­mar, clar­i­ty, tone, and orga­ni­za­tion. Chat GPT is avail­able through a brows­er or app, mak­ing it easy to access from most devices. It can adapt to var­i­ous writ­ing styles, aca­d­e­m­ic for­mats, and tone pref­er­ences. The tool can offer sug­ges­tions for sen­tence restruc­tur­ing, word choice, and over­all read­abil­i­ty pro­vid­ing expla­na­tions for edits, help­ing stu­dents improve their writ­ing skills.

How to Use Chat­G­PT for Edit­ing

Pre­pare the Text. Upload or paste your con­tent into Chat­G­PT. and pro­vide clear instruc­tions like:

  • “Check this for gram­mar and punc­tu­a­tion errors.”
  • “Sug­gest ways to improve clar­i­ty and tone.”
  • “Sim­pli­fy com­plex sen­tences with­out los­ing mean­ing.”
  • Chat­G­PT will pro­vide edits or rewrit­ten sec­tions. If need­ed, ask fol­low-up ques­tions like:
  • “Why did you sug­gest this change?”
  • “Can you rephrase this para­graph to sound more for­mal?”
  • Incor­po­rate the sug­ges­tions into your work. If a sec­tion doesn’t feel right, ask Chat­G­PT for alter­na­tive phras­ing.
  • Dou­ble-Check Accu­ra­cy:
  • Ensure fac­tu­al cor­rect­ness and prop­er cita­tions, as Chat­G­PT focus­es on lan­guage rather than con­tent accu­ra­cy.

Using Chat­G­PT as an edit­ing tool is fast and cost-effec­tive. The free ver­sion offers robust func­tion­al­i­ty, with paid options for more advanced fea­tures.

Steps to Access Chat­G­PT

  1. Vis­it Chat­G­PT  and cre­ate an account.
  2. Choose a plan – Free will work just fine
  3. Start a con­ver­sa­tion and paste your text for edit­ing!

Gram­marly sim­pli­fies gram­mar, clar­i­ty, and tone, serv­ing as a great tool for clean, pro­fes­sion­al writ­ing. To get start­ed with  Gram­marly  cre­ate an account and install the appli­ca­tion or brows­er exten­sion on your com­put­er. While you can upload entire doc­u­ments, you also have the flex­i­bil­i­ty to input spe­cif­ic sec­tions of text. This ver­sa­til­i­ty ensures you can tai­lor your edit­ing process.

Here is an exam­ple of Gram­marly in Action

Let’s edit the fol­low­ing text to demon­strate Grammarly’s capa­bil­i­ties.

Mis­pellings and gram­mat­i­cal errors can effect your cred­i­bil­i­ty. The same goes for mis­used com­mas, and oth­er types of punc­tu­a­tion . Not only will Gram­marly under­line these issues in red, it will also showed you how to cor­rect­ly write the sen­tence.

After inputting this text into Gram­marly, here is what hap­pened:

  • Red under­lines indi­cate spelling and gram­mar issues, such as “Mis­spellings” (cor­rect­ed to” Mis­spellings”) and “effect” (cor­rect­ed to “affect”).
  • Blue under­lines high­light clar­i­ty improve­ments, sug­gest­ing revi­sions to wordy phras­es or awk­ward con­struc­tions. For exam­ple, “can pos­si­bly help you revise a wordy sen­tence in an effort­less man­ner” becomes “can help you revise wordy sen­tences eas­i­ly.”

These fea­tures not only cor­rect errors but also empow­er users to refine their writ­ing for greater clar­i­ty and pre­ci­sion.

Notion­AI can be used as an edit­ing tool to refine and per­fect con­tent. It can detect and cor­rect gram­mat­i­cal errors, punc­tu­a­tion mis­takes, and spelling errors. Like Gram­marly, it will ana­lyze sen­tence struc­ture and clar­i­ty, style, and tone, and pro­vide sug­ges­tions for rephras­ing sen­tences, adding or remov­ing con­tent, and improv­ing over­all flow.

Get­ting start­ed with Notion AI is straight­for­ward. Here is a step-by-step guide to help you begin.

Step 1: Cre­ate a Notion Account . You will need an email address, Google account, or Apple ID to cre­ate an account.

Step 2: Enable Notion AI. Once you have a notion account, nav­i­gate to the Notion set­tings by click­ing on the gear icon in the top right cor­ner. Scroll down to the AI sec­tion and tog­gle the switch to enable Notion AI.

Step 3: Famil­iar­ize your­self with Notion AI fea­tures. After enabling Notion AI, explore the var­i­ous fea­tures in the fol­low­ing places:

  • Page edi­tor: Notion AI sug­ges­tions will appear as you type in the page edi­tor.
  • Block Menu: Click the three dots next to a block to access Notion AI fea­tures, such as text gen­er­a­tion and con­tent sug­ges­tions.
  • Page prop­er­ties: Some Notion AI fea­tures, like page orga­ni­za­tion and research assis­tance, can be accessed through the page prop­er­ties pan­el.

Step 4: Exper­i­ment with Notion AI fea­tures to get a feel for how they work. Try gen­er­at­ing text, get­ting con­tent sug­ges­tions, or using the research assis­tance fea­ture.

Step 5: Cus­tomize Notion AI Set­tings (Option­al). If you want to cus­tomize Notion AI set­tings, such as adjust­ing the lev­el of AI involve­ment or spec­i­fy­ing AI mod­els, you can do so in the Notion set­tings.

Exploring Notion AI as an Editing Tool

1. Famil­iar­ize Your­self with Notion AI:

  • If you haven’t already, sign up for Notion and acti­vate the Notion AI fea­ture.
  • Watch the intro­duc­to­ry tuto­r­i­al on how to use Notion AI for edit­ing. (You can find this in the Notion Help Cen­ter or through a quick YouTube search).

2. Select a Writ­ing Sam­ple:

  • Choose a short piece of writ­ing (approx­i­mate­ly 200–300 words). This could be a draft of an email, a para­graph from an essay, or a piece of cre­ative writ­ing.
  • Paste the text into a Notion page.

3. Exper­i­ment with Edit­ing Fea­tures:

  • Use Notion AI’s edit­ing tools to refine your text. Specif­i­cal­ly:
    • Gram­mar and Spelling Check: Iden­ti­fy and cor­rect errors.
    • Clar­i­ty and Style Improve­ments: Use sug­ges­tions to make your writ­ing more con­cise and pol­ished.
    • Tone Adjust­ments: Exper­i­ment with shift­ing the tone (e.g., make it more for­mal, casu­al, or pro­fes­sion­al).
    • Sum­ma­riza­tion and High­lights: Ask Notion AI to sum­ma­rize your text or high­light key points.

4. Doc­u­ment the Process:

  • Save two ver­sions of your text: the orig­i­nal and the edit­ed ver­sion using Notion AI.
  • Take note of the spe­cif­ic edits Notion AI sug­gest­ed and which ones you chose to apply or ignore.

5. Reflec­tion Ques­tions:

  • How did Notion AI improve your writ­ing?
  • Were there any sug­ges­tions that didn’t align with your intent or style?
  • Do you see your­self using Notion AI reg­u­lar­ly for edit­ing? Why or why not?
Overview

In August 2024, Marc Watkins pub­lished a sub stack called “first drafts in the AI era.” Watkins empha­sized the impor­tance of stu­dents using orig­i­nal thought to draft an out­line when writ­ing rather than rely­ing on AI to help.  He high­light­ed the val­ue  of wrestling with ini­tial thoughts and orga­niz­ing con­cepts as cru­cial to the writ­ing expe­ri­ence. Watkins was not opposed to AI; rather he argued for writ­ers to gen­er­ate their own ideas first and then use AI to enhance their work. 

Fast for­ward to today. AI tools are increas­ing­ly used as col­lab­o­ra­tive part­ners in the writ­ing process, and yes, even gen­er­at­ing ideas for that first draft. AI can brain­storm ideas, pro­vide inspi­ra­tion and gen­er­ate mul­ti­ple per­spec­tives for almost any­thing, but the human touch remains cru­cial for authen­tic­i­ty and orig­i­nal­i­ty.

In this explo­ration, we are explor­ing  Lane 3: AI as an idea gen­er­a­tor for instruc­tors in post-sec­ondary edu­ca­tion. It can be used to spark cre­ativ­i­ty and pro­vide vary­ing per­spec­tives to enhance the teach­ing and learn­ing expe­ri­ence.

Across dis­ci­plines such as health­care, math/science, and trades pro­grams, AI can gen­er­ate ideas in mean­ing­ful ways.

Healthcare

Case Stud­ies Cre­ation: Gen­er­ate com­plex, real­is­tic patient case stud­ies for sim­u­la­tion exer­cis­es, tai­lor­ing cas­es to spe­cif­ic learn­ing objec­tives (e.g., eth­i­cal dilem­mas, unusu­al symp­toms)

Exam­ples: Springer pub­lish­ing offers 7 inno­v­a­tive ways to engage a nurs­ing class with AI. Many of these ideas could be adapt­ed across oth­er dis­ci­plines.

Math/Science

Exper­i­ment Ideas: Devel­op unique lab exper­i­ments for physics, chem­istry, or biol­o­gy, such as eco­log­i­cal sim­u­la­tions or chem­i­cal reac­tion pre­dic­tions.

Exam­ple: An engi­neer­ing prof at U of Cal­gary uses AI in the class­room to sup­port stu­dents to gen­er­ate their own ideas.

Trades

Train­ing Sim­u­la­tions: Design vir­tu­al or the­o­ret­i­cal project ideas, such as cre­at­ing blue­prints for con­struc­tion or gen­er­at­ing sim­u­la­tion ideas.

Exam­ple: A trades instruc­tor could use AI to sim­u­late a haz­ardous scene and col­lab­o­rate on how best to man­age the sit­u­a­tion.

In all these cas­es, the key take­away is that AI is a tool to sup­port human cre­ativ­i­ty and crit­i­cal think­ing, not a replace­ment for it. Whether in health­care, the sci­ences, trades, or oth­er fields, instruc­tors can use AI to enhance their teach­ing, gen­er­ate new ideas, and bring their work to the next lev­el to sup­port qual­i­ty stu­dent learn­ing.

How?

Here are three AI tools that can be used to help gen­er­ate ideas beyond Chat GPT and Copi­lot:

Gam­ma AI: https://gamma.app/

Free/paid sub­scrip­tion. Can be used to gen­er­ate ideas for pre­sen­ta­tions or images. You can import files or URL’s and ask Gam­ma to enhance or gen­er­ate.

This one is a game chang­er when it comes to pre­sen­ta­tions.

When nurs­ing instruc­tor, Rachel Goodliffe, asked Gam­ma to gen­er­ate an idea for a pre­sen­ta­tion focused on inno­v­a­tive ideas in edu­ca­tion she received this:  Empowering-the-Future-Educational-Innovations.pdf

Mymap.AI: https://www.mymap.ai/idea-generator

Free/paid sub­scrip­tion. This is a fun tool that can help to gen­er­ate flow charts, time­lines, and mind maps. You can type in text, or a link/file and the tool will gen­er­ate con­tent which can then be refined but it gives you a start. Nurs­ing instruc­tor, Rachel Goodliffe, asked Mymap.AI to gen­er­ate the steps to admin­is­ter­ing an intra­mus­cu­lar med­ica­tion into the del­toid mus­cle ( she has a nurs­ing back­ground if you did not guess already). Once you input infor­ma­tion it tells you “I’m think­ing….”

While Rachel knows this is too small, she hopes you can see this impres­sive arte­fact with infor­ma­tion embed­ded. She will say that infor­ma­tion is not as com­pre­hen­sive as she would pro­vide to her stu­dents and is miss­ing safe­ty pieces such as clean­ing the skin before you inject, but as a place to begin – not bad.

https://ideanote.io/idea-generator

Idean­ote claims to be the “best online tool to quick­ly gen­er­ate ideas” and is free. It does ask you to input a com­pa­ny name, so we just added ini­tials. You can only gen­er­ate 3 ideas in 24 hours, so you must be mind­ful of that part. How­ev­er, when we asked idean­ote to help us with ideas for a project we’re work­ing on, it came up with some strong sug­ges­tions. Give it a try!

By now we’ re start­ing to think that any AI tool has capac­i­ty to gen­er­ate ideas and if that is your man­date then you are spoiled for choice. Chat GPT or Copi­lot can all gen­er­ate ideas to get you start­ed on projects or tasks, it just depends on what you need.

We decid­ed to ask Copi­lot about this:

Prompt:

Are all Gen AI resources idea gen­er­a­tors? What do you think?

Sent by Copi­lot:

Not all gen­er­a­tive AI resources are strict­ly idea gen­er­a­tors, though many do excel in that area. Gen­er­a­tive AI can serve a vari­ety of pur­pos­es, includ­ing:

  1. Con­tent Cre­ation: Writ­ing arti­cles, sto­ries, and even poet­ry.
  2. Image Gen­er­a­tion: Cre­at­ing art­work, designs, and oth­er visu­al con­tent.
  3. Data Analy­sis: Gen­er­at­ing insights from large datasets.
  4. Lan­guage Trans­la­tion: Trans­lat­ing text between dif­fer­ent lan­guages.
  5. Code Gen­er­a­tion: Assist­ing with pro­gram­ming by gen­er­at­ing code snip­pets.

So, while idea gen­er­a­tion is a sig­nif­i­cant aspect, gen­er­a­tive AI’s capa­bil­i­ties extend far beyond that.

REflect:
  1. What have you used in the class­room or for your per­son­al teach­ing  as an AI idea gen­er­a­tor?
  2. How did it work out?
Resources:

Image made with Can­va

AI is trans­form­ing the way we think about edu­ca­tion, offer­ing remark­able oppor­tu­ni­ties for per­son­al­ized learn­ing and inno­v­a­tive assess­ment meth­ods. While it is cru­cial for edu­ca­tors to estab­lish guide­lines for the respon­si­ble use of these tech­nolo­gies, it is equal­ly impor­tant for stu­dents to devel­op their abil­i­ty to engage with and crit­i­cal­ly assess the role of AI in their edu­ca­tion. By incor­po­rat­ing AI into assess­ments, edu­ca­tors can cre­ate a more engag­ing, dynam­ic, and rel­e­vant learn­ing envi­ron­ment that pre­pares stu­dents for the chal­lenges of tomor­row.

How can we effec­tive­ly encour­age stu­dents to think crit­i­cal­ly about their use of AI? To begin, it is essen­tial for edu­ca­tors to care­ful­ly con­sid­er key fac­tors when revis­ing assess­ments that facil­i­tate the inte­gra­tion of GenAI into stu­dent learn­ing. The STRIVE mod­el, devel­oped by the Uni­ver­si­ty of Cal­gary, offers a com­pre­hen­sive frame­work to guide this reflec­tion and pro­mote thought­ful engage­ment with tech­nol­o­gy:

The STRIVE val­ues — stu­dent-cen­tered­ness, trans­paren­cy, respon­si­bil­i­ty, integri­ty, valid­i­ty, and equi­ty — empha­size a holis­tic approach to the use of GenAI (see the STRIVE resource for a detailed dis­cus­sion and exam­ples). By inte­grat­ing these con­sid­er­a­tions, we can fos­ter future-focused learn­ing and skill devel­op­ment for our stu­dents, pro­vid­ing a valu­able ref­er­ence as we move for­ward with our AI assess­ment updates.

This explo­ration will focus on assess­ments in Lane 5: AI as Out­put Eval­u­at­ed from our NIC 6‑Lane AI Assess­ment Scale.

Here are six resources that offer ideas and sug­ges­tions to enhance your assess­ments:

  • AI Ped­a­gogy Project (Great col­lec­tion from Har­vard Uni­ver­si­ty – with a search tool and cat­e­gories to fil­ter by AI theme, sub­ject, skills, etc.)
  • Guide to Assess­ment in GenAI Era — Assess­ment Strate­gies (Exten­sive list of assess­ment strate­gies with sug­ges­tions for avoid­ing inap­pro­pri­ate use vs. respon­si­ble prac­tices for GenAI from the Uni­ver­si­ty of Water­loo)
  • Assess­ment Design using GenAI (Dis­cus­sions on GenAI assess­ment design with spe­cif­ic UBC course exam­ples found at the bot­tom of page in the drop down: How to make the most of GenAI in assess­ments and assign­ments — UBC exam­ples)
  • Assess­ment Design — GenAI-inte­grat­ed Assess­ments (From the Uni­ver­si­ty of Alber­ta – spe­cif­ic course exam­ples under Cat­e­go­ry 3 — Explore Exam­ples)
  • GenAI — AI Out­put Eval­u­at­ed (BCIT list of ideas from mul­ti­ple dis­ci­plines for assess­ments that crit­i­cal­ly eval­u­ate AI out­puts)
  • Dis­ci­pline-spe­cif­ic exam­ples of effec­tive prompt writ­ing (Uni­ver­si­ty of Cal­gary prompt library with many great ideas for stu­dent prompts and course assess­ments with some dis­ci­pline spe­cif­ic exam­ples near the bot­tom of the page)

As you review the resources, which ideas stand out to you as par­tic­u­lar­ly rel­e­vant for enhanc­ing assess­ments in your course?

Explore:

  • Choose one of the assess­ment ideas from the resources above (or choose anoth­er idea you have either cre­at­ed or applied)
  • Reflect on and plan how you might inte­grate it into your course or a future course.

Addi­tion­al Resources:

• The STRIVE mod­el (Uni­ver­si­ty of Cal­gary)
• Gen­er­al Assess­ment Design Strate­gies (Uni­ver­si­ty of Water­loo)
• Teach Any­where: Prompts, Tutors and Assis­tants

Through the adop­tion of a 6‑Lane AI Assess­ment Scale, NIC has strate­gi­cal­ly moved to pro­mote the inte­gra­tion of AI tools. Thought­ful­ly used, these tools have the poten­tial to assist learn­ers in ways that we are only begin­ning to under­stand.

Since we are explor­ing the sixth and final lane, Full AI, in the assess­ment scale, it will be valu­able to high­light a few quotes that con­tex­tu­al­ize where edu­ca­tion stands with AI and offer some can­did reflec­tions on what is no longer fea­si­ble in this quick­ly chang­ing land­scape.

“…by redesign­ing our assign­ments to include in-class ele­ments, direct con­nec­tions to local and cur­rent issues, and a focus on the process in addi­tion to the prod­uct, we can make GenAI use eas­i­er to detect and mit­i­gate when it is not appro­pri­ate for learn­ing goals.”

UBC AI in Teach­ing & Learn­ing

“It is a poten­tial­ly dis­rup­tive but absolute­ly nec­es­sary per­spec­tive which under­stands that per­mit­ting any use of AI effec­tive­ly per­mits all use of AI, and since it is unde­tectable and sophis­ti­cat­ed across domains, the dis­tinc­tion between pre­vi­ous Lev­els 2, 3, 4, and even 5 is some­what arbi­trary.”

Leon Furze (blog post)

“…we take the stance that assess­ment secu­ri­ty, in the tra­di­tion­al sense, is only pos­si­ble at Lev­el 1, but that assess­ment valid­i­ty is pos­si­ble through­out.”

Leon Furze (blog post)

“I think gen­er­a­tive AI is unavoid­able, not inevitable. The for­mer speaks to the real­i­ty of our moment, while the lat­ter address­es the hype used to mar­ket the promise of the technology—a sales pitch and lit­tle else. Fac­ul­ty and stu­dents have to con­tend with gen­er­a­tive tech­nol­o­gy in our world as it is now, not as it is promised to be. That should be our focus.

Marc Watkins (Sub­stack)

To con­clude, this final para­graph helps put things in per­spec­tive that we are not alone in the edu­ca­tion field in our wrestling with AI and its unknown impacts. Read the arti­cle, engage with the Empa­thet­ic Voice Inter­face, then reflect on our chang­ing future.

“The major­i­ty of OpenAI’s 250 mil­lion week­ly users are stu­dents. Ope­nAI and oth­er gen­er­a­tive com­pa­nies have com­mit­ted to release ver­sions of their tools for free with few safe­guards, in a mas­sive pub­lic exper­i­ment that defies belief. There is no touch­stone moment in edu­ca­tion­al his­to­ry that com­pares to our cur­rent AI moment. If you think gen­er­a­tive AI is like MOOCs, then I invite you to have a three-minute dis­cus­sion about that very top­ic with a mul­ti­modal AI tool called EVI—Hume.AI’s Empa­thet­ic Voice Inter­face. You don’t even need an account, sim­ply click the link and pick the syn­thet­ic per­sona of your choice. Get emo­tion­al with it and see how quick­ly it responds to match your mood. Do you still believe this tech­nol­o­gy won’t pro­found­ly change edu­ca­tion, labor, or even soci­ety itself?”

(empha­sis added to all quotes)

This explo­ration will focus on Lane 6 of the assess­ment scale: Full AI. It is impor­tant to under­stand that this lane is reserved for skills that are assess­able out­side the con­text of AI. To phrase this anoth­er way, the skills that are being eval­u­at­ed would be unaf­fect­ed by the stu­dents use of AI.

Below are 4 different options to get you and your students moving with Full AI assessment integration:
  1. Have stu­dents cre­ate a slide deck with AI for a class pre­sen­ta­tion that they have pre­vi­ous­ly researched and stud­ied
    In this sit­u­a­tion, the pre­sen­ta­tion itself is not eval­u­at­ed but the con­tent and find­ings are. Stu­dents may need to present the infor­ma­tion from pre­vi­ous stud­ies but lack the skills nec­es­sary to com­plete this in a man­ner that is visu­al or func­tion­al.
  1. Stu­dents work with AI to cre­ate an orig­i­nal ad cam­paign
    In this sit­u­a­tion stu­dents could con­verse with the AI to estab­lish a qual­i­ty end prod­uct by con­tin­u­al eval­u­a­tion. Since AI will like­ly be an avail­able tool to a novice ad cre­ator in the work­place, it would be rea­son­able for them to prac­tice with it in the class­room envi­ron­ment.
  1. Have stu­dents cre­ate AI-gen­er­at­ed mar­ket analy­sis reports, includ­ing pre­dic­tive mod­el­ing of mar­ket trends using AI for data analy­sis and visu­al­iza­tion.
    In this sit­u­a­tion, the stu­dent is not eval­u­at­ed on the analy­sis reports or visu­al­iza­tions but rather on their inter­pre­ta­tions of the data pro­vid­ed and their pro­pos­als for mit­i­ga­tion strate­gies. Each stu­dent would create/generate their own unique data set. Here are exam­ples of such a case study with Chat GPT
  1. Have stu­dents use AI for the cre­ation of visu­al designs, logos, and brand­ing mate­ri­als, focus­ing on the refine­ment and appli­ca­tion of design prin­ci­ples to AI-gen­er­at­ed con­cepts.
Reflect:
  • Con­sid­er how you could inte­grate AI ful­ly into a project or activ­i­ty in your course
  • What are your thoughts? Does this work?
Resources: