Emergency Preparedness

Will It Transform Higher Education?

I recently had a conversation with a colleague who stated this eye-popping fact about using AI for research papers: “In a class of 18, I have four students at 100% and one more at 61%. The rest were 0%.”

She was referring to the artificial intelligence detection score for a writing assignment given in an undergraduate-level STEM course. Sadly, this exact conversation happens every day here at the University and every other educational institution, according to The Guardian.

Getting students to avoid using AI for research papers and other assignments is a daily reality for most of us. For teachers with large-enrollment courses, the scope of the problem is overwhelming and exhausting.

The frustration is palpable and justified. We are concerned about what the use of AI by students means for learning and for the extra time we must spend in policing their work.

We instructors are not police, nor do we want to be. The more time we spend policing whether students are using AI for research papers, the less time we spend educating them. It’s a lose-lose situation.

Technology Aids vs. Education

I was in elementary school when calculators began entering classrooms in the 1970s, and their use was forbidden. By the time I was in high school, we were allowed to use calculators in very limited, controlled and directed ways.

Most teachers were concerned that calculators would prevent us from building computational skills, but others saw calculators as beneficial tools that could be used to aid teaching and enhance understanding. These two views, however, are not mutually exclusive.

By the time I entered graduate school in the 2000s, the same debate about the use of statistical software had been settled. Students in the life sciences were required to take a two-course series in statistics.

In the first course, we learned statistics fundamentals and underlying theory. In the second course, we learned how to appropriately use SAS, the statistical software of choice at my institution.

The first course was critical. Without the basic knowledge it provided, we would not have been able to properly vet the output of the statistical software.

I also took graduate courses on multivariate statistics and ordination, both of which were software-based. In fact, those methods weren’t even possible prior to computing. They existed only in theory.

It is the same story for Bayesian statistics. Those changes were major advancements in my field.   

Research shows that the appropriate and guided use of calculators and other technologies in math courses can increase learning and accessibility. The National Council of Teachers of Mathematics includes this statement as one of its Principles and Standards for School Mathematics: “Technology is essential in teaching and learning mathematics; it influences the mathematics that is taught and enhances students’ learning….Technology can help support investigation by students in every area of mathematics and allow them to focus on decision making, reflection, reasoning, and problem solving.”

Today, the American Statistical Association offers statistics teachers peer-reviewed lesson plans for grades 6-8 using statistical analysis software. The American Statistical Association’s Guidelines for Assessment and Instruction in Statistics Education includes “Use technology to explore concepts and analyze data” as one of its recommendations.

The Detection of AI for Research Papers Is a Continuous Arms Race

Current and emerging technologies claim to detect AI-generated content. However, workaround solutions are rapidly being launched.

There are already tools out there such as Aithor.com whose tagline is “Your undetectable AI writer.” With new options for creating papers with undetectable AI-generated content arising every day, we can’t rely on detection as our only solution. As with most situations, the best defense is a good offense

Instructors Need to Move Fast

We are not without recourse for preventing the use of AI for research papers. But as with any fundamental shift in education, teachers must be open to change.

Gone are the days where student assessments can primarily consist of fact-based essays focused on finding and synthesizing existing information. Gone are the days where discussions are based solely on writing to show an understanding of information.

We must emphasize more doing, creating, analyzing and connecting to classroom material. Achieving these goals will mean reimagining our courses and reminding ourselves of exactly what it is we want our students to know and be able to do.

What skills will they need when they seek opportunities to enter the workforce? How can we make assessments more practical and deter students from having AI complete assignments for them?

Embracing Technology as an Instructional Tool

As with any technology, there will always be a risk of student misuse with AI. There will always be those students who take the shortcut, even at the cost of building their personal knowledge and skills. Instructors will still have the responsibility of calling out and creating consequences for those students who misuse technology.

If AI is used with proper acknowledgement, encouragement, and training, our students will surely benefit from it. The use of AI will increase students’ learning and give them important professional skills, and AI’s misuse will decrease over time.

Source link

Related Articles

Back to top button
  • bitcoinBitcoin (BTC) $ 66,242.00 1.1%
  • ethereumEthereum (ETH) $ 3,069.85 1.64%
  • tetherTether (USDT) $ 0.999797 0.04%
  • bnbBNB (BNB) $ 574.88 0.69%
  • solanaSolana (SOL) $ 170.49 1%
  • usd-coinUSDC (USDC) $ 0.999670 0.02%
  • staked-etherLido Staked Ether (STETH) $ 3,068.81 1.59%
  • xrpXRP (XRP) $ 0.509006 2.23%
  • the-open-networkToncoin (TON) $ 6.27 1.97%
  • dogecoinDogecoin (DOGE) $ 0.148912 2.63%
  • cardanoCardano (ADA) $ 0.468694 2.67%
  • shiba-inuShiba Inu (SHIB) $ 0.000024 3.85%
  • avalanche-2Avalanche (AVAX) $ 35.69 3.83%
  • tronTRON (TRX) $ 0.121241 1.46%
  • wrapped-bitcoinWrapped Bitcoin (WBTC) $ 66,307.00 1.05%
  • chainlinkChainlink (LINK) $ 16.58 2%
  • bitcoin-cashBitcoin Cash (BCH) $ 484.50 1.97%
  • polkadotPolkadot (DOT) $ 6.94 2.94%
  • nearNEAR Protocol (NEAR) $ 7.82 1.09%
  • matic-networkPolygon (MATIC) $ 0.684123 3.39%
  • litecoinLitecoin (LTC) $ 82.18 1.96%
  • internet-computerInternet Computer (ICP) $ 12.60 4.82%
  • uniswapUniswap (UNI) $ 7.64 2.86%
  • fetch-aiFetch.ai (FET) $ 2.20 3.36%
  • daiDai (DAI) $ 0.998885 0.23%
  • leo-tokenLEO Token (LEO) $ 5.89 0.05%
  • ethereum-classicEthereum Classic (ETC) $ 27.61 3.45%
  • hedera-hashgraphHedera (HBAR) $ 0.111097 3.07%
  • render-tokenRender (RNDR) $ 10.04 2.59%
  • pepePepe (PEPE) $ 0.000009 3.92%
  • wrapped-eethWrapped eETH (WEETH) $ 3,189.21 1.54%
  • first-digital-usdFirst Digital USD (FDUSD) $ 0.999093 0.1%
  • aptosAptos (APT) $ 8.11 5.27%
  • immutable-xImmutable (IMX) $ 2.24 5.34%
  • crypto-com-chainCronos (CRO) $ 0.121697 2.64%
  • cosmosCosmos Hub (ATOM) $ 8.21 4.64%
  • arweaveArweave (AR) $ 47.72 1.77%
  • filecoinFilecoin (FIL) $ 5.54 5%
  • mantleMantle (MNT) $ 0.939204 3.98%
  • stellarStellar (XLM) $ 0.105544 2.27%
  • renzo-restaked-ethRenzo Restaked ETH (EZETH) $ 3,031.65 1.16%
  • okbOKB (OKB) $ 49.03 2.29%
  • kaspaKaspa (KAS) $ 0.122173 1.08%
  • blockstackStacks (STX) $ 1.96 4.67%
  • the-graphThe Graph (GRT) $ 0.300144 3.62%
  • optimismOptimism (OP) $ 2.45 3.98%
  • arbitrumArbitrum (ARB) $ 0.973201 4.42%
  • makerMaker (MKR) $ 2,772.90 1.8%
  • dogwifcoindogwifhat (WIF) $ 2.52 1.11%
  • vechainVeChain (VET) $ 0.034308 2.72%