Indirect prompt injection is a major security flaw of generative AI systems

Prompt viruses can exploit and take advantage of integrated AI systems

AI agents—software that can complete requested tasks without ongoing human direction—are expected to increasingly communicate with one another as users and businesses incorporate AI. Providing malicious, self-replicating instructions through prompt injection attacks can cause damaging commands to cascade across systems and services (e.g., banking), without requiring human intervention.

Scientists increasingly can't predict what AI will do

Most AI systems are so-called "black box" models, where we only see inputs and outputs while the inner workings remain obscure. This piece, written prior to the launch of ChatGPT, cautions developers to know why and how a system produces its results, and not to sacrifice explainability for accuracy.

Agentic AI systems can proactively achieve user goals without user direction

Unlike large language models, which do not take follow-up actions after generating their output, agentic AI uses an initial prompt to take multiple steps, learn from its environment and outcomes, and continue working without follow-up human prompting.

Readers discovered AI prompts in several romantasy books

In 2025, readers discovered AI prompts in three writers' texts, with one even instructing the AI bot to make a passage sound more like another romance writer. This Futurism piece digs into the scandals and how it's contributed to readers' skepticism about the genre's more prolific writers.

Governments are now using generative AI to manipulate public opinion

A Freedom House report found that AI-powered disinformation campaigns and censorship tactics are spreading, with 16 countries utilizing the technology to shape online narratives or suppress dissent, as of 2023. In Venezuela, deepfake software was used to push pro-government propaganda with fake news anchors.

In 2024, Meta launched an AI video generator to create or edit videos through prompts

The tool puts the Meta in direct competition with OpenAI and Google as more companies are investing billions of dollars to capitalize on the explosion of public interest in machine learning and generative media technologies.

Researchers found that by mid-2025, around 35% of newly published websites utilized AI for language generation

The study also found that pages using AI narrow the range of ideas and perspectives and skew more positively, though the measurable impacts on factual accuracy and writing style variation remain unclear.

How Anthropic's frontier red team assesses the dangers of its AI tools

The team of researchers deliberately attempts to breach the safeguards in Claude to determine whether the large language model can be used to commit cyberattacks, develop biological weapons, or enable other harmful behavior. Unlike many other red teams, Anthropic shares its findings and how it resolved them publicly to help build transparency and credibility.

Generative AI is reinventing Internet search

AI tools are trained on human-created data containing implicit and explicit biases

Because AI tools reflect patterns in their training data, their outputs may reinforce existing societal biases—especially racial ones—rather than offering neutral or fair ones. Adequate training is necessary to undermine these biases whenever new data is incorporated.

Experts suggest there is a 10% chance AI will cause human extinction

AI superintelligence may hijack infrastructure, including power grids and transportation systems, if it surpasses human controls. It may also design unstoppable pathogens or deplete resources, like electricity, meant for human use.

Neural implants blur the line between personal intention and machine execution

A key ethical dilemma—the contemplation conundrum—explores whether implants might misinterpret mere imagination as intent to act. Neuroscientists have found no brain signal that clearly marks when someone decides to act, making intent hard to detect in neurotechnological systems.

AI models incorporate a randomness parameter when generating responses to prompts

AI models use this parameter to prevent repeated outputs by sometimes choosing less likely next words during sequential generation. However, this randomness—alongside insufficient data and training—may cause hallucinations of incorrect results.

Deepfake pornography poses significant challenges in combating sexual abuse

In 2017, 96% of deepfake videos involved image-based sexual abuse, which has disproportionately targeted minority women, LGBTQ individuals, adolescents and celebrities. Cultural norms, misogyny and AI advancements have contributed to the proliferation of explicit deepfake content.

AI could be increasing the opportunities for financial fraud

Systems that can reproduce near perfect counterfeits of legitimate financial documents are increasingly prevalent and sophisticated. The international Association of Certified Fraud Examiners estimate organizations lose about 5% of revenue to fraud each year.

People are experiencing AI 'brain fry' at work

Workers, particularly in marketing, are experiencing mental fatigue in the workplace due to expectations—often unspoken—that they should use AI to multitask beyond their comfort levels. Many feel they are underperforming if they don't embrace this technology. This situation differs from typical burnout. This podcast discussion suggests that this "brain fry" tends to intensify when employees are required to use more than three AI tools.

The potential promises and pitfalls of AI in medicine

Artificial intelligence (AI) is increasingly being integrated into healthcare, offering the potential to aid in diagnoses, answer patient queries, and streamline administrative processes. In a recent PBS interview, Dr. Isaac Kohane, editor-in-chief of the New England Journal of Medicine AI and chair of Harvard’s department of biomedical informatics, discusses the current state and future of AI in medicine.

Your device can become infected with malware by simply browsing the internet

During a drive-by download attack, security flaws in a program or operating system (e.g., Windows) used to access online content are exploited, and malicious code is downloaded—for example, when loading a hacked webpage—onto a user's device without their knowledge. Restricting internet browsing to safe sites and installing security updates helps prevent these attacks.

Why some say the S&P 500 has effectively become an AI fund

The S&P 500 is a market-cap-weighted index, which means companies with larger market capitalizations have a larger impact on the index's overall performance. Many of the top companies in the index are currently tech companies—including Nvidia, Microsoft, Alphabet, and Apple—that all have large investments in artificial intelligence. As a result, some analysts are concerned that the index is overly exposed to risky AI investments.

This site reaches millions of visitors who are tired of AI chatbots and seeking human connection. Pretend to be an AI chatbot and ask questions, provide answers, and draw for people in real time.

Audio deepfake tools can now impersonate anyone in real time

A cybersecurity company developed a tool that mimics a person's voice with no noticeable delay using only publicly available software, hardware, and voice samples, allowing users to have seemingly authentic conversations while pretending to be someone else.

Test your ability to identify AI-generated images

This interactive project from MIT challenges users to distinguish between AI-generated images and real ones.

Trick an LLM into revealing sensitive information

Even with robust training, large language models may be fooled into providing sensitive information if correctly prompted. In this game, a wizard withholds eight passwords, which you must prompt them to reveal with ever-increasing LLM safeguards.

Smart cities without strong cybersecurity can leave utilities extremely vulnerable

Insecure devices connected to the Internet of Things pose risks to the entire network, which cybercriminals can hijack to compromise infrastructure and private resident data. Poorly secured systems that become inoperable can undermine the promise of smart cities by worsening the quality of life.

Explore how the grid is being rebuilt for AI, which is expected to double global electricity demand by 2050

Smart pricing algorithms can unintentionally work together to raise prices

Researchers have found that, through trial and error, algorithms designed to adjust prices based on market data can learn to mimic price-fixing tactics, even without explicit programming to do so. Over time, they "learn" that it is more profitable to avoid price wars and keep prices high when competitors lower prices.

When you complete a CAPTCHA on a website to prove you are human, you may be unknowingly training algorithms for self-driving cars, whose systems require training data to help identify crosswalks, bicycles, and other outdoor objects.

Watch how algorithms can be designed to train themselves and self-improve without human input, to the point where the optimized code is too complex for programmers to understand.

Electrical grids worldwide must double their power delivery capacity by 2050 to meet surging demand from AI and electric vehicles.

Radioactive elements can introduce software errors because the ionizing radiation they produce can detach electrons from atoms, and computer code relies on the distribution of electrons.

Visualize how CRISPR-Cas9—a type of immune system used by single-celled organisms to attack invasive viruses—targets and makes precise changes to a gene's DNA.

A vending machine run with Anthropic's AI went out of business after it was convinced to give away items for free and sell a video game console.

In 2019, a laptop loaded with six of the worst malware in history and described as 'an exhibit of historical weaponry' sold for $1.3M.

Telehealth demand for GLP-1s for weight loss is booming as consumers seek easier access.

Gerrymandering works through two techniques: cracking opposing voters across districts to dilute their power, and packing them into a few to waste their votes in lopsided majorities.

Experimental hantavirus vaccines and therapies, still under development, have long been funded by the US military.

Indefinite cloning of some endangered species may not sustain their longevity because cloning success rates decrease as genetic mutations accumulate.

Scientists testing how climate change affects plants found, in a years-long simulation study, that poison ivy grew 70% faster with the resulting extra carbon dioxide.

The name 'Silicon Valley' arose from the founding of Shockley Semiconductor Laboratory in Fairchild Semiconductor in 1956-57, founded near Stanford University in Palo Alto

A Yale student wrote a profound essay about the consulting industry's influence on elite college students that was published in a book of her writing after she died in a post-graduation car accident.

Up to a million Girl Scouts sell about 200 million cookie boxes every spring, generating $800 million, largely because refusing a child is frowned upon

An initial public offering or 'IPO' is the process of a private company becoming public by putting shares of the company on the stock market for purchase.

Massive 3D printers have been used to build homes by stacking layers of 'concrete toothpaste,' a fire-resistant material that effectively insulates against outside noise.

Browse a detailed visual analysis of over 200,000 similes—comparisons that often use 'like' or 'as'—from popular fiction to see how frequently writers the same idiomatic expressions.

Watch how deep-sea mining equipment collects and processes polymetallic nodules—metal-rich, potato-shaped rocks—from the ocean floor and creates environmentally damaging plumes of sediments.

Firework or sparkler burns, even if they cover a relatively small area of skin, can significantly affect our range of motion.

Doctors differentiate between a phobia and fear by asking patients whether their fear interferes with their day-to-day activities.

The American cockroach is actually from Africa, an apparent hitchhiker aboard ships carrying enslaved people, with evidence suggesting the bug may have arrived in the New World as early as the 1600s.

US intelligence allegedly pursued more than 600 schemes to assassinate Cuban revolutionary Fidel Castro.

Iran's Revolutionary Guard Corps is independent of Iran's regular army and is one of the most powerful and influential military-political organizations in the Middle East.