- Everyday AI
- Posts
- Is ChatGPT Making College Better or Worse?
Is ChatGPT Making College Better or Worse?
Google delays Project Astra, ChatGPT in education, $50B investment on AI power, AMDās AI struggles and more.
Everyday AI
October 30, 2024
š Subscribe Here | š£ Hire Us To Speak | š¤ Partner with Us | š¤ Grow with GenAI
Outsmart The Future
Today in Everyday AI
7 minute read
š Daily Podcast Episode: Although itās been a few years since ChatGPT was released, many college kids are still using ChatGPT in their own ways. So it makes us think, is ChatGPT making college better or worse? Give it a listen.
šµļøāāļø Fresh Finds: Googleās code is more than 25% AI-generated, Apple to add AI reviews in App Store and BBC reports on social media networks profiting off AI misinformation. Read on for Fresh Finds.
š Byte Sized Daily AI News: Google delays Project Astra to 2025, AMDās AI struggles and a $50 billion investment for AI power. For that and more, read on for Byte Sized News.
š AI In 5: Hereās a sleeper feature inside Perplexity thatāll save you an insane amount of time. See it here
š§ Learn & Leveraging AI: We break down how ChatGPT and other AI tools can benefit college students to prepare them for the workforce. Keep reading for that!
ā©ļø Donāt miss out: Did you miss our last newsletter? We talked about how to use Claude's agentic AI, Salesforce launching Agentforce, GitHub partnering with Google and Anthropic, U.S. setting restrictions on AI in China and LinkedIn unveiling an AI hiring assistant. Check it here!
Is ChatGPT Making College Better or Worse? š§āšļø
Most college kids aren't actually learning AI.
They're just using ChatGPT to write their papers.
Problem? Yes.
Is there a solution? Well, there's some good next steps, and Jason Gulya.
Join the conversation and ask Jordan and Jason questions on AI in education here.
Also on the pod today:
ā¢ AI in Education Over Time ā°ļø
ā¢ AI's Impact on Employability š¼
ā¢ Institutional Approach to AI Adoption š«
Itāll be worth your 40 minutes:
Listen on our site:
Click to listen
Subscribe and listen on your favorite podcast platform
Listen on:
Hereās our favorite AI finds from across the web:
New AI Tool Spotlight ā Phase is an open-source application manager, Contacted is AI powered content and SEO reports and MinusX is AI data scientist for your analytical apps.
OpenAI ā OpenAI has rolled out two new updates to its API and has also open-sourced a new benchmark called SimpleQA.
Two Realtime API updates:
- You can now build speech-to-speech experiences with five new voicesāwhich are much more expressive and steerable. š¤£š¤«š¤Ŗ
- We're lowering the price by using prompt caching. Cached text inputs are discounted 50% and cached audio inputs are discountedā¦ x.com/i/web/status/1ā¦
ā OpenAI Developers (@OpenAIDevs)
4:51 PM ā¢ Oct 30, 2024
Google ā During Googleās Q3 Earnings call, it was announced that more than a quarter of new code at Google is generated by AI.
Google has also released a new prompting essentials AI course.
Social Media ā Redditās AI push is starting to pay off as it saw a 68% revenue increase in Q3.
A new report from BBC shows how social media networks are profiting from AI misinformation.
Apple - Apple is rumored to be adding a new auto-generated review summary for apps in the App Store.
Business of AI - Eleven Labs has hired the team behind Omnivore, an open source reading app.
AI in Education ā MITās College of Computing has launched a postdoctoral program to advance AI across disciplines.
Read This ā Ashton Kutcherās VC firm, Sound Ventures, is investing in Fei-Fei Liās World Labs.
1. Google Delays Project Astra AI Agent Launch Until 2025 š
In a recent earnings call, Google CEO Sundar Pichai confirmed that the highly anticipated Project Astra, aimed at creating AI apps capable of real-time, multimodal understanding, won't hit the market until 2025 at the earliest.
Initially expected to roll out a consumer-focused agent experience by December, this timeline shift raises questions about the future of AI integration in everyday tasks like shopping and travel planning. The project showcases promising technologies, such as smartphone apps that can recognize surroundings and engage in contextually relevant conversations.
2. Alphabet's AI Ambitions Shine in Q3 Earnings Call āØ
In its latest earnings call, Alphabet CEO Sundar Pichai emphasized the companyās steadfast focus on artificial intelligence, asserting that their innovative strategies are driving user engagement across platforms like Google and YouTube. While advertising revenues showed some weakness, the company reported a staggering 14-fold increase in API usage for its AI model, Gemini, with over a billion users benefiting from enhanced search functionalities.
As Alphabet navigates legal challenges regarding monopoly claims, Pichai remains optimistic about the companyās competitive edge in AI, promising that their investments will not only boost user experience but also unlock new monetization avenues.
3. $50 Billion Bet on AI Power Needs š
KKR and Energy Capital Partners have joined forces to invest a staggering $50 billion in data-center and power-generation projects, aiming to address the soaring energy demands of artificial intelligence over the next four years. As tech giants scramble for electricity to fuel their data centers, this partnership highlights the urgent need for reliable energy sources amid growing constraints on the U.S. power grid.
With natural gas positioned as a key player in meeting these demands, tech companies are caught in a dilemma between their fossil fuel reliance and sustainability goals.
4. AMD's AI Sales Outlook Disappoints, Shares Slide š
Advanced Micro Devices (AMD) faced a rough start this Wednesday as its fourth-quarter revenue forecast of $7.5 billion fell short of Wall Street's expectations of $7.55 billion, signaling slower-than-expected growth in AI sales. While the chipmaker anticipates over $5 billion in revenue from AI accelerators this year, investors were hoping for a more significant increase.
CEO Lisa Su reassured stakeholders about AMD's growing presence in the market, highlighting customer openness to their offerings despite ongoing supply challenges. Following the announcement, AMD shares dropped nearly 9% in premarket trading.
5. U.S. DOD's AI Commitment Takes Stage at Forum š”
In a timely address at the Responsible Artificial Intelligence in Defense Forum, Deputy Defense Secretary Kathleen Hicks underscored the Pentagon's unwavering commitment to ethical AI practices, highlighting the importance of responsible military use. With nearly 60 nations backing the Political Declaration on Responsible Military Use of Artificial Intelligence, Hicks emphasized that strong ethical standards set the U.S. apart from its competitors.
The Defense Department's proactive approach includes updated policies and practical resources now available to international allies and tech firms, reinforcing a collective effort toward trustworthy AI.
6. SAG-AFTRA Strikes Groundbreaking Deal with Ethovox š£
In a significant development for voice performers, SAG-AFTRA has struck a deal with Ethovox to ensure fair compensation for artists whose voices train a new AI model aimed at creating digital replicas. This agreement not only secures session fees and ongoing revenue sharing but also emphasizes diversity and inclusion in the voice model's development.
As the voice acting community continues to navigate challenges, including an ongoing strike by video game voice actors, this contract sets a precedent in the industry, promising to protect and empower performers.
STOP Ignoring This Perplexity Feature! Save Hours Daily
So many people are sleeping on this ONE feature inside Perplexity thatāll save you an insane amount of time.
Weāre showing you Perplexity Collections and how to use it.
Check out today's AI in 5.
š¦¾How You Can Leverage:
Itās been two DANG years since ChatGPT dropped.
Yet, ChatGPT in the classroom is still a polarizing topic.
Why?
Sure, some forward-thinking universities are tackling AI in education the right way.
But like 90% of colleges are straight up fumbling the ball before even taking the field.
(Likeā¦ yāall are blanket banning ChatGPT on college campuses? Thatās embarrassing. You do know almost every student is using AI and your AI ācontent detectorsā do not actuallyā¦. Detect AI content.)
Jason Gulya joined Everyday AI to help provide some clarity.
Jason wears many hats when it comes to AI in education. Heās a professor, consultant to educational institutions, and also advises Berkeley College on AI implementation.
So we sat down to answer the question ā is ChatGPT making college better or worse?
Jasonās answer was two-fold.
He says ChatGPT has made the student experience slightly better, but that colleges have got to make some HUGE changes quickly before the whole experience gets much worse.
(Big agree.)
Whether youāre an educator, student or have kids in the classroom, todayās show is a must watch/listen/read.
Ready for the Cliff Notes?
Letās get it.
1 ā The rules are confusing šµāš«
Jason can empathize with the average student.
They could be taking 6 different classes, all with wildly different AI policies. When Universities give leeway to educators, it often means a complete pile of WTF for students.
See, when ChatGPT exploded onto the scene two years ago, almost every college and university started with a blanket ban.
But then something VERY obvious happened.
Employers wanted new hires with AI readiness, and colleges werenāt teaching it.
Try this:
Confusing, right?
We agree.
Jason had some advice for educators that we think students and parents can take to their classrooms.
Transparency and communication.
If AI is allowed, it needs to be communicated. Jason talked about using a 5-point scale, with zero being AI not being allowed and 5 being where itās a requirement and a backbone to the assignment.
When we talk about AI and use it correctly, we can cut through the confusion and improve the learning experience.
2 ā All software is AI now š§āš»ļø
Jason told us a pretty common story in higher ed right now.
AI is banned, but software thatās viewed as educational is encouraged.
So many say no to ChatGPT, but yes to tools like Grammarly.
Jason said that teachers who had banned ChatGPT yet encouraged Grammarly were in for a LOL awakening when they realized that Grammarly essentially has a ChatGPT mode inside of it.
Try this:
Get educated, educators.
(This is us talking now, not Jason.)
First things first ā AI content detectors literally do not work. Likeā¦. theyāre about as useless as air conditioners in Antarctica.
OpenAI even shut its own AI content detector down after it was shown to only be 26% accurate.
(And we only think they created it as a marketing ploy, TBH.)
Second, your students are smarter than you when it comes to using AI tools. Once you ban LLM Tool A, theyāve already moved on to LLM Tool B and C.
You canāt play whack a mole with software banning.
Educators need to understand that just about every piece of software has AI baked into it by default.
Grammarly.
Google Docs.
Microsoft Word.
Heckā¦. WINDOWS! The entire operating system!
So trying to still actively ban AI makes educators look uneducated.
/rant over/
Oh, while youāre here and weāre on this topic, itās worth checking out this fantastic related episode with Jules White on the future of AI in the classroom.
3 ā Process over product ā
Jason dropped a heavy gem on our heads toward the end of the show.
The college process is all wrong.
And AI has made us all realize that.
He said if nothing else, the proliferation of AI tools has put a giant mirror in front of collegeās respective faces.
Like, are we really basing like 80% of a studentās college experience based on obnoxiously long writing assignments?
(That students are 100% gonna spit through AI anyways.)
Jason said the cultureās gotta change. And fast.
Right now, AI in higher ed is a mess. So, colleges gotta lean into that mess to fix it.
Or risk becoming irrelevant.
(That might sound crazy, but watch it happen.)
Try this:
Writing papers should probably be a very small part of the college experience in the age of AI.
Or just get rid of em.
Jasonās found a happy medium.
Heās using AI in the classroom to actually change the learning experience. From personalized learning GPTs to contrarian AI roleplaying that makes students defend their positioning, ChatGPT is actually a bonkers good learning tool.
Go check our last episode with Jason, where he shared some more of his inside secrets on practical ways to use AI in the classroom.
ā
Numbers to watch
1 Billion
According to Alphabetās Q3 report, more than 1 billion global Gemini users are using Gemini as part of the search function.
Now This ā¦
Let us know your thoughts!
Vote to see live results
What's your favorite part of the newsletter? |
Reply