AI will match human creativity for sure
Tribune News Service
Most of the academic writers I know acknowledge that artificial intelligence is beginning to encroach on their field, including the ability to write a cogent essay demonstrating critical thought. If it's a bit bland or generic, well, so are a lot of student essays that we've read over the years. But we still claim as uniquely human the ability to write creatively. Only humans can combine feelings and words and their own associations to form art. That's not true. AI is, among other things, a web scraper, a gatherer of what's out there, a synthesizer. But so are writers, and what we value is their idiosyncratic skill at putting together what's out there. As poet T.S. Eliot wrote in his essay 'The Metaphysical Poets': 'When a poet's mind is perfectly equipped for its work, it is constantly amalgamating disparate experiences; the ordinary man's experience is chaotic, irregular, fragmentary. The latter falls in love, or reads Spinoza, and these two experiences have nothing to do with each other, or with the noise of the typewriter or the smell of cooking; in the mind of the poet these experiences are always forming new wholes.'
But AI can assemble something akin to that, even if the formation currently involves more fumbled tries in its blind workings toward complexity than a human writer might attempt. In fact, the processes of a large language model, or LLM, are slightly opaque to even computer scientists. And the models are trained on huge sets of data, as are humans. The students in my writing workshops strive for something original, a high bar to pass. Definitions of originality vary widely: presumably something we haven't seen before. In fact, a lot of us need to read more because what we think of as original has been done and done before we were born. 'There is nothing new under the sun,' reads a famous line from Ecclesiastes, undoubtedly not the first time that sentiment has been expressed in a similar way.
So if originality, per se, doesn't exist, what are we looking for? This: Take a known element, such as a toddler's birthday party, and join it to an element you usually don't see associated with it, such as a funeral. The occasion becomes decidedly mixed. I just made up that scenario, but you can too. It's called recombinant. The video player is currently playing an ad. Or take an ordinary object, such as a pen, and in your story, use it to dig a hole. This is repurposing. And lest you scoff at such an easy method to produce something new, understand that we certainly do applaud that as something original, from innovative ad campaigns to a short story that includes five languages.
Those who judge creative writing in the classroom are so interested in process — 'This poem went through five drafts' — that we forget that in the end, the finished poem is what matters most. Does it make a difference whether a short story is the result of months of painful choices and revising or simply randomized selections that finally yield a masterpiece (six monkeys at typewriters eventually coming up with Hamlet)?
If so, why? A creative writer produces a work of art. Are the 'how' and 'why' relevant? How does one know, anyway? Think of the Turing test: If an observer can't tell whether a response is human-generated or machine-generated, then the machine can be said to exhibit intelligence. And if some discerning critics still fondly believe that the novel they admire could have been created only by some demented genius in a garret or basement or castle, that's romantic delusion. Perhaps it was written by the AI inhabiting a data server in Texas.
Last year, a poet in a nearby English department thundered that AI would never write poetry like Emily Dickinson. So another professor typed into ChatGPT, 'Write a quatrain in the manner of Emily Dickinson that uses black-and-white imagery and the idea of drowning, yet with a sense of elation.' In a matter of seconds, the AI had executed a pretty darned good quatrain, which I won't quote because you can try it yourself, and anyway, ChatGPT will come up with a new one every time.
The kicker? The thundering poet termed the result uninspired and banal, so the other professor went into Dickinson's 1,800-poem corpus, retrieved an obscure quatrain and presented it to the poet, who called the result banal and uninspired. A surprisingly high number of professors I've talked to have strong opinions on AI but have never tried it out.
Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles
Gulf Today
05-05-2025
- Gulf Today
To dumbly go where no space budget has gone before
Christopher Cokinos, Tribune News Service Reports that the White House may propose nearly a 50% cut to NASA's Science Mission Directorate are both mind-boggling and, if true, nothing short of disastrous. To make those cuts happen — a total of $3.6 billion — NASA would have to close the Goddard Space Flight Center in Maryland, and cancel the mission that will bring back samples of Mars, a mission to Venus and the Nancy Grace Roman Space Telescope, which is nearly ready to launch. Every space telescope besides the Hubble and the James Webb would be shut down. According to the American Astronomical Society, some cuts would include projects that help us understand the sun's effects on global communications, a potential national security threat. Casey Dreier, the policy advocate for the Pasadena-based Planetary Society, says, 'This is an extinction-level event for the Earth- and space-science communities, upending decades of work and tens of billions in taxpayers' investment.' In addition, NASA as a whole would see a 20% cut — just as we are moving forward with the Artemis program. Artemis is NASA's step-by-step 'Moon to Mars' human spaceflight campaign. Artemis II is set to launch sometime next year and will send four astronauts on a lunar fly-by, the first time humans have been in close proximity to another celestial body in more than 50 years. While it seems likely that Artemis will continue in some fashion, a 20% overall agency budget cut won't leave any part of NASA unaffected. The president promised a 'golden age of America'; his nominee to head NASA promised a 'golden age of science and discovery.' This would be a return to the dark ages. Taking a blowtorch to space science would also have little effect on the federal budget while setting back American leadership in space — and the inspiration it provides across political divides — by generations. The Astronomical Society warns that our cutbacks will outsource talent 'to other countries that are increasing their investments in facilities and workforce development.' And, as Dreier points out, spacecraft would be 'left to tumble aimlessly in space' and billions wasted that have already been spent. 'Thousands of bright students across the country,' he wrote recently, 'would be denied careers in science and engineering absent the fellowships and research funds to support them.' Here's the dollars-and-cents context. NASA's budget since the 1970s 'hovers' between 1% and 0.4% of the federal discretionary spending, according to the Planetary Society's analysis, yet for every dollar spent, NASA generates $3 in the national economy. NASA's giveback was worth nearly $76 billion in economic impact in 2023, supporting more than 300,000 jobs. In California alone, NASA and its associated partners in industry and academia provide more than 66,000 jobs, more than $18 billion in economic activity and $1 billion in state tax revenue. NASA's bang-for-the-buck is astronomical, pun intended. Cutting waste is one thing. Evisceration is another. When it comes to science — from public health to climate change — the current administration is doing the latter, not the former. Meanwhile, China continues its space ambitions, with plans for a human lunar campaign and its own 'sample return' mission to the Red Planet. For now, fortunately, the bipartisan support for NASA seems to be holding. Democrats and Republicans in Congress, led by the Planetary Science Caucus, have spoken out against this attack on NASA. And the Planetary Society has engaged thousands of passionate activists to fight this battle. Humans yearn for connection to the universe — so we watch launches on social media, we follow the tracks of rovers on Mars and we marvel at creation in pictures transmitted from the James Webb Space Telescope. We borrow telescopes from the public library and look to the heavens. Bending metal — the actual process of making rovers and spaceships and telescopes — drives economic activity. Fascinating results — the data from space science missions — fires the imagination. We choose to go to space — sending humans and probes — and we pursue knowledge because curiosity is our evolutionary heritage. We explore other worlds to know them and, in doing so, we discover more about ourselves. If you agree, let Congress know. That may be the only backstop against dumbly going where no budget has gone before.
Gulf Today
03-03-2025
- Gulf Today
AI will match human creativity for sure
Tribune News Service Most of the academic writers I know acknowledge that artificial intelligence is beginning to encroach on their field, including the ability to write a cogent essay demonstrating critical thought. If it's a bit bland or generic, well, so are a lot of student essays that we've read over the years. But we still claim as uniquely human the ability to write creatively. Only humans can combine feelings and words and their own associations to form art. That's not true. AI is, among other things, a web scraper, a gatherer of what's out there, a synthesizer. But so are writers, and what we value is their idiosyncratic skill at putting together what's out there. As poet T.S. Eliot wrote in his essay 'The Metaphysical Poets': 'When a poet's mind is perfectly equipped for its work, it is constantly amalgamating disparate experiences; the ordinary man's experience is chaotic, irregular, fragmentary. The latter falls in love, or reads Spinoza, and these two experiences have nothing to do with each other, or with the noise of the typewriter or the smell of cooking; in the mind of the poet these experiences are always forming new wholes.' But AI can assemble something akin to that, even if the formation currently involves more fumbled tries in its blind workings toward complexity than a human writer might attempt. In fact, the processes of a large language model, or LLM, are slightly opaque to even computer scientists. And the models are trained on huge sets of data, as are humans. The students in my writing workshops strive for something original, a high bar to pass. Definitions of originality vary widely: presumably something we haven't seen before. In fact, a lot of us need to read more because what we think of as original has been done and done before we were born. 'There is nothing new under the sun,' reads a famous line from Ecclesiastes, undoubtedly not the first time that sentiment has been expressed in a similar way. So if originality, per se, doesn't exist, what are we looking for? This: Take a known element, such as a toddler's birthday party, and join it to an element you usually don't see associated with it, such as a funeral. The occasion becomes decidedly mixed. I just made up that scenario, but you can too. It's called recombinant. The video player is currently playing an ad. Or take an ordinary object, such as a pen, and in your story, use it to dig a hole. This is repurposing. And lest you scoff at such an easy method to produce something new, understand that we certainly do applaud that as something original, from innovative ad campaigns to a short story that includes five languages. Those who judge creative writing in the classroom are so interested in process — 'This poem went through five drafts' — that we forget that in the end, the finished poem is what matters most. Does it make a difference whether a short story is the result of months of painful choices and revising or simply randomized selections that finally yield a masterpiece (six monkeys at typewriters eventually coming up with Hamlet)? If so, why? A creative writer produces a work of art. Are the 'how' and 'why' relevant? How does one know, anyway? Think of the Turing test: If an observer can't tell whether a response is human-generated or machine-generated, then the machine can be said to exhibit intelligence. And if some discerning critics still fondly believe that the novel they admire could have been created only by some demented genius in a garret or basement or castle, that's romantic delusion. Perhaps it was written by the AI inhabiting a data server in Texas. Last year, a poet in a nearby English department thundered that AI would never write poetry like Emily Dickinson. So another professor typed into ChatGPT, 'Write a quatrain in the manner of Emily Dickinson that uses black-and-white imagery and the idea of drowning, yet with a sense of elation.' In a matter of seconds, the AI had executed a pretty darned good quatrain, which I won't quote because you can try it yourself, and anyway, ChatGPT will come up with a new one every time. The kicker? The thundering poet termed the result uninspired and banal, so the other professor went into Dickinson's 1,800-poem corpus, retrieved an obscure quatrain and presented it to the poet, who called the result banal and uninspired. A surprisingly high number of professors I've talked to have strong opinions on AI but have never tried it out.
Gulf Today
27-01-2025
- Gulf Today
We are in a mental health crisis, a ‘moonshot' is needed
Tribune News Service On New Year's Day, Master Sgt. Matthew Livelsberger reportedly blew himself up in a Tesla Cybertruck in an apparent suicide outside the Trump International Hotel in Las Vegas. Despite having a successful career as a Green Beret and being the father of a baby born just last year, Livelsberger reportedly suffered from severe post-traumatic stress disorder and depression. This is just one high-profile example of how depression and anxiety disorders, ineffectively treated, can have lethal consequences. Rates of PTSD combined with anxiety disorders or diagnoses of depression increased by 177% and 138% in only five years, from 2019 to 2023, according to the most recent edition of the Medical Surveillance Monthly Report, a peer-reviewed journal of the Armed Forces. Skyrocketing rates of depression and anxiety are also being seen in civilian adults and, alarmingly, in our nation's children. According to a recent behaviour risk survey from the Centers for Disease Control and Prevention, almost 40% of children in ninth through 12th grade experienced persistent feelings of sadness and hopelessness, and 20.4% seriously considered attempting suicide. Compounding this crisis in mental health is our inability to treat depression and anxiety effectively in most people. Selective serotonin reuptake inhibitors, such as Prozac, Zoloft and Paxil, the first line of medical therapy for these mood disorders, induce symptom improvement in only half of those people taking them and place only a third into remission. Unintended side effects of SSRIs, including anxiety itself, suggest that SSRIs may actually cause some of the symptoms they are meant to reverse. In pregnant women, SSRIs cross the placenta and are absorbed into breast milk; they may affect fetal development and cause disorders of gut-brain interaction. In short, we need better therapies for depression and anxiety, and we need them as soon as possible. Disturbingly, however, the budget provided for mental health research by the National Institutes of Health, the nation's premier source of funding for medical research, actually decreased last year. Emerging research in mental health is showing how little we know and thus how much work needs to be done, including in how SSRIs work. SSRIs increase serotonin, which has been called the 'happy hormone' due to its anti-depressive and anti-anxiety effects. Since Prozac made a splash in the 1980s, the general scientific consensus has been that SSRIs enter the bloodstream and help anxiety and depression by increasing serotonin directly in the brain. While serotonin is found in the brain, more than 90% of our bodies' serotonin actually is found in the gut. Furthermore, SSRIs are systemically absorbed, leading to an increase in serotonin levels in not only the brain but also the gut. This leads to a key question — when SSRIs help anxiety and depression, is that because they increase the serotonin in the brain, or because they increase the serotonin in the gut? In other words, might it be better to target anxiety and depression by focusing on the gut, thus increasing efficacy of the SSRIs and reducing their harmful side effects? I recently published a study in animals that shows that increasing serotonin specifically in the gastrointestinal tract improves anxiety and depression. This suggests that a medication that can increase serotonin in the gut, while not entering the bloodstream, may be a more effective treatment for anxiety and depression while also avoiding fetal exposure in women taking SSRIs. (Pregnant women should not stop taking SSRIs when their physicians think they are required, as maternal depression itself can have its own negative effects.) It is critical, however, that studies focusing on how the gut affects mental health be at the forefront of political and scientific policies. With a new presidential administration, there must be a key focus on prioritising mental health initiatives. The formation of the bipartisan Senate Mental Health Caucus has been a very positive start. An even more robust effort, similar to the White House initiative 'Cancer Moonshot,' would be required. This revolutionary initiative is building new collaborations and programs across all parts of the cancer community, including the federal government, private companies, health care providers, patient groups, philanthropies and the general public. Many facets of Cancer Moonshot could be implemented in the mental health sphere. These could include investments in widespread screening and preventative programs; access to quality, coordinated care; the implementation of navigation services to enable patients to more easily acquire services; and prioritized research funding directed toward new treatments and preventative measures. Importantly, all of these measures consider equity for all populations and also consider military members and children specifically.
