Scientists propose nuclear plan to stop Earth-bound asteroids

By Syeda Samar Zahra Zaidi When you ask an AI chatbot a question it does not know the answer to, an odd thing can happen: instead of hesitating or saying, “I am not sure!” it confidently delivers a tidy, well-structured, completely fabricated response—often with fake citations, fictional figures, and an authoritative tone. This phenomenon is known as the ‘hallucination trap ’, and it may be the most significant barrier between trust and artificial intelligence. I was introduced to the term “AI hallucination” for the first time in the workshop under the project titled “Digital Trust and Free Expression in the AI Era” at Department of Media and Development Communication, University of the Punjab, supported by the U.S. Mission to Pakistan in partnership with the Pakistan-U.S. Alumni Network (PUAN). I learned that a “hallucination” occurs when a language model produces information that sounds convincing but is factually incorrect or completely made up. Examples include inventing a legal case that never existed, citing a research paper that doesn’t exist or boldly misrepresenting a historical date. What is disturbing is not merely that AI can be wrong-humans make mistake too- but rather the way it presents that incorrect information with clearly, confidence and without a trace of doubt. Large language models are essentially highly complex pattern predictors trained to produce the most likely next word in a sequence. As a result, when a model has encountered a topic thousands of times in its training data, the predictors are typically accurate. However, when it comes across a gap, an unidentified fact or a question at the outer edge of its knowledge, it doesn’t shift into a hesitant mode. Instead, it continues doing exactly what it was trained to do: generate a logical, fluent piece of writing, producing a response that is structurally sound and grammatically correct but based on nothing. In human communication, confidence is generally perceived as a sign of strength which is important. We naturally prefer to trust a calm, reliable voice over a hesitant one. AI systems unintentionally exploit the tendency because their language remains confident whether they are right or wrong. The risks increase with the use of AI, as students may unintentionally include false information into their work and journalists have published articles citing artificially created sources and quotes, negatively affecting the credibility of their publications. Such inaccuracies influence what millions of people see, believe or share, thereby amplifying fabricated content. The reason AI does not simply say “I don’t know” is partly due to the way these systems are trained. Many models are designed to provide helpful, comprehensive-sounding answers and are rarely rewarded for expressing doubt during training. This means that a model that frequently acknowledges uncertainty may be judged as less helpful than one that consistently provides an answer, even when that confidence is occasionally misplaced. Such incorrectly aligned incentives arise when the training process gradually teaches the model that sounding confident is more valuable than being accuracy. Retrieval-augmented generation, which relies on responses in actual, verifiable records rather than internal memory alone, improved training techniques that explicitly account for uncertainty; and systems built to simply cite sources. So, users can independently verify claims are just a few of the ways developers are using to addresses this problem. While none of these solutions fully eliminate hallucination yet, taken together they represent a major shift from blind confidence toward earned trust. The most dangerous outcome is not an AI that makes mistakes but a culture of users who stop questioning because the system always sounds so sure of itself.   Therefore, until AI becomes more reliably aware of it own limitation, part of the responsibility falls on the people who use it: treating AI-generated output as a starting point rather than an endpoint, cross checking important facts against credible sources. In the end, the hallucination trap serves as a reminder that illusion is not the same as truth. As AI becomes increasingly common in education, digital media and everyday life, understanding why it occasionally lies with confidence is not only a technical curiosity but also a fundamental aspect of digital literacy. We must strengthen our ability to verify the information provided by AI as they become more persuasive. The views, thoughts, and opinions expressed in this blog are solely those of the author and do not necessarily reflect the official policy or position of the U.S. Mission to Pakistan or USEFP. The writer, Syeda Samar Zahra Zaidi, is a student of Media and Development Communication at  University of the Punjab, Lahore and can be reached at samarzahra456@gmail.com.  

Previously, I believed that if one could see something, it could be trusted because seeing is believing. A video was proof. A voice recording was evidence. That belief is gone, and I don’t think we have fully sat with the implications. There is an abnormal shift happening. AI-generated images, voice cloning, face-swapping apps, and similar tools were supposed to expand how people express themselves, but instead they have slowly chipped away at what matters most: the belief that people can express themselves through their images and their words. I attended a workshop under the project titled “Digital Trust and Free Expression in the AI Era” at Department of Media and Development Communication, University of the Punjab, supported by the U.S. Mission to Pakistan in partnership with the Pakistan-U.S. Alumni Network (PUAN) last week. During the discussions on digital trust. I took part in the discussion and shared my thoughts that we are now in a time where we must take this out of the realm of fun tech novelty and place it in the realm of a structural issue in how societies operate. Everyone says deepfakes are a “misinformation” issue, but this is not an adequate way to describe it. Misinformation is something that’s a false statement that can, in theory, be fact-checked. Deepfakes strike at the heart of video and audio credibility which have always been treated as vivid reality. When people realize that any clip can be fabricated, they stop trusting all clips, even real ones. This is what concerns me most. Fake things are believed and real things become deniable. Consider the implications of journalism, courts, political accountability, etc. If a politician is recorded saying something damaging, they can simply claim that it’s a “deepfake” and, depending on how much trust citizen have lost, people may believe them. We are creating a society in which it is possible to negotiate authenticity and in which negotiable authenticity is a gift for those who wish to avoid consequences. I also face a conflict here: I strongly believe in free speech, satire, parody, art and provocation even with synthetic media. The same technology behind de-aged actors, resurrected voices, dubbing, and experimental AI is extremely creative, yet it is all connected to malicious deepfakes. There is no way to ‘clean’ a tool of its worst use-case. A hammer can break windows or build houses. I do not think banning synthetic media is the answer, nor do I support pre-emptive censorship. That path harms legitimate satire, journalism and art under the pretense of ‘preserving truth’. History does not look kindly on institutions that decide what is a “fair” fake. Freedom of expression has never been of consequence or friction, and right now, deepfakes have almost no friction. The reality is there is no cost, disclosure, and accountability in using this technology deceptively. The issue is the imbalance. I believe the solution is not reducing synthetic tools but improving provenance. We need a digital watermarking, which cannot be “easily removed”: cryptographic signatures embedded into cameras, microphones and generative tools at the point of creation, so that the “unverified” label appears by default, not as an afterthought. Not to prohibit any content but to allow people to know what they are seeing before deciding whether to trust it. Platforms have been allowed to operate without responsibility for too long. Convincing fakes can reach millions before correction and corrections never travel as far. This asymmetry is not natural; it is a design choice. Platforms can mark, rate or tag synthetic content at the time of dissemination. Most simply haven’t had to! But, beyond platforms, we must change our own instincts as individuals. The old saying was “pics or it didn’t happen.” The new rule should be: “Who gains from believing this and can I verify it myself? A society that believes everything or worse, believes nothing is in danger. The deepfake age is not the age of AI; it is the consequences of a world where the cost of faking reality is near zero and the cost of establishing truth remains high. Freedom of expression should protect the most challenging and profound exchange of ideas, not the production of fake consent, fake evidence, or fake people. We can have creative freedom and still expect honesty about sources. These are not competing prioritie These are not different objectives/priorities. However, if we pretend it’s a content-moderation footnote, we must make sure we are not waking up in a world where “fake it till you make it” has become ‘fake it till you break it” and by that time, it won’t be just one video or one lie that will break. It will be our fundamental ability to reach a shared understanding of reality. The views, thoughts, and opinions expressed in this blog are solely those of the author and do not necessarily reflect the official policy or position of the U.S. Mission to Pakistan or USEFP/PUAN. The writer, Mahnoor Ashraf Wattoo, is a student of Media and Development Communication at University of the Punjab, Lahore 

Since Ashura, the mothers of Sindh have been sitting in at Babarloi Bypass through suffocating heat and humidity, with one demand: return our missing and abducted children. For a mother, a child is her entire universe — and for years these women have lived in psychological anguish searching for theirs. Fazila Sarki of Thul vanished nearly seventeen years ago; her father named the local Wadero he says orchestrated it, yet she was never recovered. Priya Kumari of Sangrar is entering her fifth year missing — committees formed, updates promised, and still nothing, as if the earth or sky simply swallowed her. Sindh’s landlords and chieftains have built a parallel state, one that operates above the law. If not, why would fourteen children currently be missing, why would the soil be stained with young blood over trifles, why would mothers go blind from weeping and cemeteries keep expanding? Innocent young women are still declared Kari and honor-killed — denied even a ritual bath, shroud, or funeral prayer, sometimes dumped in rivers. The law exists on paper; hundreds of thousands of officials draw salaries and perks from public taxes to enforce it — yet ordinary people have no protection at any level. The papers are full of these stories daily. The law exists, but it is locked away in books. Are the hands meant to enforce it helpless, or simply indifferent? Sukkur Civil Hospital — the teaching hospital for Ghulam Muhammad Mahar Medical College, and once a lifeline stretching from Balochistan’s Sibi, Nasirabad, and Bhag Nari to Kandhkot-Kashmore, Shikarpur, and Ghotki — has sunk to the level of Shikarpur’s Rai Bahadur Udhodas Hospital. Department heads sit in name only; real control rests with a handful of favorites who have run these institutions into the ground. Road accidents are a daily occurrence on the highways, and while the Orthopaedic Department stays busy, its facilities inspire only regret. With Nasir Shah and veteran politician Khurshid Shah both present, this hospital ought to be a model for Northern Sindh — instead it is anything but. Law and order, and justice, are the pillars any civilized society rests on. Where they hold, life, property, and honor are protected, and social harmony follows. Where they fail, fear, distrust, and unrest take their place — and our own society is proof enough of that; little more needs saying. An educated person lifts not just himself but his society; where ignorance and social ills grow instead, it is education that has failed first. Healthcare here tells the same story: women give birth in Qingqi rickshaws, medicine is scarce, and families buy even a disposable syringe out of pocket — though the budget for it exists. Doctors draw government salaries while absent, some working in Qatar or Saudi Arabia, others running private clinics. A friend recently told me PPHI, on taking over a government health unit, found a stockpile of medicines for heart disease, diabetes, and stomach ailments — worth a fortune, and only months from expiry. What tyranny is greater than this: medicine sitting unused while the needy go without? It is nothing short of an atrocity against the public. Hunger, disease, unemployment, lawlessness, and ignorance have battered Sindh until its people’s capacity for critical thought is paralyzed. The state, along with conscious citizens, politicians, and the well-off, must turn to this crisis — urgently.