Photo credit: Macrumors
Apple just put Active Noise Cancellation, Adaptive Audio, and stem volume swipes into the open ear pair that costs $149, and that price is the whole story. AirPods 5 with Wireless Charging Case arrived in stores on September 18 at the same starting money Apple used for AirPods 4 two years ago, except ANC is now standard instead of a $50 upcharge. You still get the familiar short stem and the same 4.3 gram bud that rests in the bowl of the ear with no silicone tip. A redesigned multiport acoustic layout and next generation Adaptive EQ are meant to add depth and detail across a wider range of ear shapes, while Personalized Spatial Audio with dynamic head tracking keeps music, movies, and games moving with your head.

For $149, you get all of the things that the cheaper $129 USB-C variant lacks. There’s a force sensor on the stem that allows you to play, pause, and switch between listening modes with a squeeze, and a swipe up or down provides you volume control without having to pull out your phone. The case can be charged via USB-C, any old Qi pad, or even an Apple Watch charger, and it even has a built-in speaker so Find My can ring it if it decides to hide beneath the couch. Apple claims that this device would last 5 hours with ANC, 7 hours without, and 22 hours with the cover. 5 minutes in the case will give you an extra hour, which is one hour every charge and two hours more total than the cheaper model.

Apple AirPods 5 with Wireless Charging Case

• ACTIVE NOISE CANCELLATION—AirPods 5 offer improved Active Noise Cancellation to help reduce outside noise.
• HEAR THE WORLD AROUND YOU—Adaptive Audio blends Active Noise Cancellation and Transparency for the best listening experience in any environment…
• LIVE TRANSLATION—Communicate across languages using Live Translation, enabled by Apple Intelligence.*

Open ears are surprisingly comfortable for long periods of time and still let some ambient noise to pass through even when the ANC is turned on, which is the trade-off for not using foam tips. Apple claims that these buds can block out up to 50% more outside noise than the AirPods 4 with ANC, making it the most powerful open ear cancelation they’ve ever done. Adaptive Audio simply blends the cancelation with the Transparency mode to make it sound natural in shifting surroundings. Transparency is about making nearby voices sound exactly like they do in real life. Then there’s Conversation Awareness, which simply turns down the audio when you start talking and back up when you finish. To top it all off, dual beamforming mics and Voice Isolation ensure that your calls are clear even in the wind or in heavy traffic. In addition, the same circuitry can record studio-quality voice notes in case you need to scribble something down, and they’re IP57 rated for durability, so you can use them in the rain or during a workout without worry.

When it comes to software, the “similar to Pro” pitch becomes very effective. As long as you have a compatible iPhone and the most recent software, Live Translation allows you to converse with someone and translate it on the run with a stem press or a Siri request. New Siri AI, rolling out in English with iOS 27 on Apple Intelligence phones, lets you ask longer questions and handle more tasks without touching glass. A nod or a small head shake can accept a call or dismiss a prompt.

Talks have begun between US and Chinese officials to propose a mechanism for the two countries to notify each other of artificial intelligence incidents which could threaten national security.

The framework, called the US China AI Dialogue, would also see representatives from both countries meet again to discuss the technology and try to align on common goals and threats, Treasury Secretary Scott Bessent said.

“We think that, just like with any cross-border activity, that moving from opaque to more transparency between the number one and the number two AI powers in the world is very important,” Bessent told reporters on Sunday.

Bessent said the discussions picked up from where they had been left off after US President Donald Trump’s visit to Beijing in May. Since then, public safety concerns about the technology have grown. Security incidents over the summer, including OpenAI’s agents hacking Hugging Face, highlighted the capabilities of frontier models and how the industry’s safeguards have failed to keep pace.

Fears reached a fever pitch earlier this month after Anthropic employees warned AI could be on course to wipe out humanity. In recent weeks, leaders of the biggest frontier laboratories, including OpenAI’s Sam Altman, Anthropic’s Dario Amodei, and Elon Musk, have called for a slowdown in development of the technology while they set safety standards. In an essay suggesting steps for laboratories and governments to pace the technology’s development, Amodei had asked for government assistance in establishing international coordination, including with the Chinese Communist Party.

Trump has so far mostly brushed off their calls, despite growing bipartisan support for regulating the industry. The US government already has sufficient criminal and regulatory power over developers, his administration has argued.

He’s not alone. Jensen Huang, chief executive officer of Nvidia, which builds AI infrastructure, pointed out that AI firms are already subject to cybersecurity and damage liability laws. “Apply that first,” he said in an interview with CBS News over the weekend. “Don’t let this doomsday narrative allow someone to relieve them of the laws that currently exist.”

Trump has said any slowdown could cede the US lead in the technology to China—a danger because, in his own words: “WHOEVER WINS AI, WINS!” His chief science advisor David Sacks previously warned that “China is very unlikely to join a global agreement” on pacing the technology. Thomas Wolf, cofounder of open-source AI platform Hugging Face, which Nvidia recently agreed to buy, has questioned whether the US can build global cooperation while “explicitly stating you want to design it to keep widening your own lead.” “That seems like a pretty counterproductive way to start the conversation to me,” he wrote on X in response to Amodei’s essay.

The US is currently trying to temper China’s advance in AI by restricting exports of advanced chips and chipmaking technology to China. Officials did not discuss these US export controls in this weekend’s meetings, said Bessent, which also featured trade discussions and were “very successful” ahead of this week’s visit to Washington by Chinese President Xi Jinping.

Spoofing means falsifying or imitating an identity, while phishing is deception designed to make you reveal information, send money, open malicious content, or take another unsafe action. Smishing and vishing are phishing variants delivered through text messages and voice communication respectively.

Spoofing vs Phishing vs Smishing vs Vishing at a Glance

The four terms are closely related, but they do not describe four mutually exclusive attack categories. Spoofing primarily describes false identity or source information, while phishing describes the deception itself. Smishing and vishing narrow phishing by communication channel.

Four Threats Compared

The most important difference is therefore not simply email versus text versus phone. Spoofing describes how identity can be falsified, while phishing describes how a person is manipulated. Smishing and vishing identify two channels through which that manipulation can occur.

The Key Relationship: Spoofing Is a Technique, While Smishing and Vishing Are Channels of Phishing

The Federal Bureau of Investigation defines spoofing as disguising information such as an email address, sender name, phone number, or website URL so communication appears to come from a trusted source. The same FBI guidance explains that phishing schemes often use spoofing techniques to make their bait more convincing.

That gives the terms different jobs. Spoofing answers a question about identity: what source is being imitated or falsified? Phishing answers a question about deception: what is the attacker trying to persuade the victim to do?

Smishing and vishing narrow the discussion further by describing the delivery channel. The FBI identifies smishing as phishing through SMS text messages and vishing as phishing through phone calls, voice email, or Voice over Internet Protocol communications.

These categories can therefore overlap. Imagine an attacker who pretends to be your bank. The attacker could spoof the bank’s telephone number, send you a fraudulent text message, call you after you respond, and then direct you to a fake login page. The same campaign could involve spoofing, smishing, vishing, and phishing without contradiction.

This relationship also explains why treating all four terms as interchangeable can cause confusion. Phishing belongs to the broader category of phishing and social engineering, while spoofing can also describe forms of impersonation outside a phishing scenario.

What Is Spoofing?

Spoofing is the act of making an identity, source, or piece of identifying information appear to be something it is not. In communications-related attacks, the false information might be a caller-ID number, email sender name, email address, domain, or website URL.

For example, your phone might display the genuine customer-service number of your bank even though the call came from an attacker. The Federal Trade Commission warns that scammers can make any name or number appear on caller ID. The displayed number is therefore not proof of who is actually speaking.

Email and web spoofing can work in similar ways. An attacker may create a look-alike domain, alter a character in an address, or design a fake website to resemble a trusted service. In September 2025, the FBI warned that criminals were creating spoofed versions of the Internet Crime Complaint Center website by altering characteristics of the legitimate domain.

This article uses “spoofing” mainly in the communications and social-engineering sense. The technical term is broader. NIST’s cybersecurity glossary includes several spoofing definitions, including impersonating an authorized user and manipulating data or signals. The exact meaning therefore depends on technical context.

A compromised legitimate account is another important edge case. If an attacker sends a malicious message from a real employee’s hijacked mailbox, the visible account may be genuine even though the person controlling it is not. That differs from simply forging visible sender information, but both situations can make fraudulent communication appear trustworthy.

What Is Phishing?

Phishing is a social-engineering attack that tries to persuade someone to disclose sensitive information or perform an action that benefits the attacker. NIST’s phishing glossary includes definitions centered on fraudulent solicitation, deceptive electronic communication, and attempts to trick people into revealing sensitive information.

A phishing message might ask you to:

• enter a password on a fake login page;
• provide banking or card details;
• open a malicious attachment;
• approve an unexpected login or multifactor authentication request;
• send money or change payment instructions; or
• provide confidential personal or business information.

For example, an email could claim that your bank account has been locked and instruct you to sign in immediately. The link leads to a page designed to resemble the bank’s real website. The spoofed branding or URL helps establish trust, while the phishing element is the attempt to manipulate you into entering credentials.

The word “phishing” is often associated with email because email is a common delivery route. However, the broader phishing family is not limited to one channel. The FBI explicitly describes smishing and vishing as variations that use similar deception through text and voice communication.

What Is Smishing?

Smishing is phishing conducted through text messaging. The name combines SMS, or Short Message Service, with phishing. The FBI defines smishing as malicious targeting through SMS or Multimedia Messaging Service text messages.

A smishing message might claim to be:

• a bank fraud alert;
• a package-delivery problem;
• an unpaid toll or fee;
• a job opportunity;
• a message from a government agency; or
• a person who supposedly contacted the wrong number.

The objective is not always an immediate malicious download. A scammer may first want you to reply, establish a conversation, move to another messaging platform, or trust a later request. FTC guidance on scam and phishing text messages documents fake account alerts, package notices, malicious links, and attempts to steal passwords, account numbers, and other personal information.

A May 2025 FBI warning documented attackers sending text messages while impersonating senior U.S. officials, then attempting to establish rapport or move targets to another communication platform. The FBI described the text-messaging stage as smishing.

The channel is what makes the phishing attempt smishing. A fake bank alert sent by email is not smishing merely because the underlying story is identical to one used in a fraudulent text.

What Is Vishing?

Vishing is voice-based phishing. The term combines voice with phishing. The FBI’s general phishing guidance includes phone calls, voice email, and Voice over Internet Protocol (VoIP) calls as vishing channels.

A typical vishing attack might involve someone claiming to be from a bank, government agency, technical-support department, employer, or other trusted organization. The caller may create urgency by saying money is disappearing from an account, taxes are overdue, a computer is compromised, or an immediate security check is required.

The attacker may then ask for a password, banking information, security code, payment, remote computer access, or another action. The FTC warns that scammers use false identities, urgency, and other stories to obtain money or personal information through fraudulent phone calls.

Artificial intelligence can make this form of impersonation more convincing. In May 2025, the FBI documented AI-generated voice messages used in vishing campaigns impersonating senior U.S. officials. A December 2025 update reported that malicious actors continued using text and AI-generated voice messages in the campaign.

Voice cloning does not create a separate fifth category alongside phishing, smishing, and vishing. It is a technique that can strengthen a vishing attack by making an impersonated voice more believable.

How the Same Scam Can Use All Four Techniques

Consider a bank-impersonation campaign. It could begin when an attacker sends a text claiming that a suspicious payment has appeared on your account. The text contains the bank’s name and asks you to confirm whether you recognize the transaction.

That first contact is smishing because the phishing attempt arrived through a text message. If the sender information, branding, or website is deliberately made to imitate the bank, spoofing is also involved.

The attacker might then call you from a number that appears to match the bank’s real customer-service line. That false caller ID is another form of spoofing, while the spoken attempt to make you reveal a verification code or move money is vishing.

Finally, the caller might direct you to a fake bank website where you are asked to enter your username and password. The broader attempt to manipulate you into surrendering those credentials is phishing.

Trying to label the entire campaign with only one of the four terms loses useful information. A security team may need to know both the attacker’s social-engineering objective and the separate channels and impersonation techniques used to carry it out.

Warning Signs That Apply Across All Four

The delivery channel changes, but many social-engineering warning signs remain similar. None of these signs proves fraud by itself, but several appearing together should make you slow down and verify the request independently.

• Unexpected urgency: You are told to act immediately before you have time to verify the story.
• Threats or fear: The sender or caller claims your account, money, job, benefits, or legal status is at immediate risk.
• Requests for passwords or security codes: Someone asks for credentials, one-time codes, or approval of an unexpected authentication request.
• Unusual payment instructions: You are told to send cryptocurrency, gift cards, wire transfers, or money to a new destination.
• Suspicious links or domains: A web address resembles a legitimate service but uses altered spelling, extra words, or a different domain.
• Pressure to stay within the attacker’s communication path: The person discourages you from hanging up, checking an official app, contacting a colleague, or verifying the request elsewhere.
• A sudden request to move platforms: A text, email, or voice message asks you to continue the conversation on another messaging service.
• Identity details that cannot be independently confirmed: The message looks familiar, but the claimed sender cannot verify the request through a trusted channel you already know.

Legitimate messages can sometimes be urgent, contain links, or arrive from unfamiliar numbers. The safer decision is therefore based on independent verification rather than one visual clue or one warning sign.

If you have already clicked, sent money, disclosed credentials, or granted device access, the problem has moved beyond identifying the attack type. The appropriate recovery depends on what the attacker obtained. The practical steps for dealing with spoofed calls and emails after contact include protecting affected accounts, contacting payment providers where necessary, checking exposed devices, and preserving useful evidence.

The Difference That Matters Most

The easiest way to keep the four terms straight is to ask different questions. Spoofing asks which identity or source is being falsified. Phishing asks what deception is being used to make the target act. Smishing and vishing identify text and voice as the channels carrying that phishing attempt.

Because those dimensions can overlap, one scam can legitimately fit more than one label. Understanding the relationship is more useful than forcing every attack into a single category.

The big picture: Cornell University researchers have developed a way to recycle lithium-ion batteries without breaking them down into raw materials first. Their method, called Direct Electrode-to-Electrode Regeneration, or DEER, restores worn battery electrodes so they can be used again. The results suggest that some EV batteries could be refurbished at lower cost and with less energy than conventional recycling methods, though the process will not work for every degraded cell.

The work focuses on a common problem in aging lithium-ion batteries: buildup on the electrodes. As batteries are charged and discharged, a layer known as the solid electrolyte interphase, or SEI, grows on their surfaces. Some SEI is necessary, but too much of it slows the movement of electrons and reduces battery capacity.

In the Cornell process, researchers take apart used cells and place the electrodes in an electrochemical bath containing 1,3-dimethyl-2-imidazolidinone, or DMI. The treatment removes the thicker SEI layer without destroying the electrode itself. The renewed electrodes can then be used to build new battery cells.

The researchers reported that regenerated cells recovered up to 95% of their original capacity. They also found that treated batteries degraded more slowly than untreated ones. Untreated degraded batteries lost capacity at a rate of 0.072% per cycle, compared with 0.042% per cycle for batteries restored through DEER. The lower degradation rate held for about 800 cycles before increasing.

A second round of treatment brought a previously regenerated battery back to 90% of its original capacity. That suggests the process could extend the working life of electrodes more than once, depending on the condition of the battery.

The approach differs from most commercial recycling methods. Conventional systems typically shred or crush spent batteries into a mixture called black mass. Recyclers then use high heat, chemicals, or both to recover lithium, cobalt, nickel, manganese, copper, and aluminum.

Those methods recover valuable materials, but they also destroy electrodes and other parts that may still be usable. DEER treats the battery less as a source of metals and more as a device with components that can be repaired.

That could be useful for electric-vehicle batteries. EV battery packs are often removed from vehicles once they fall to about 70% to 80% of their original capacity. At that point, the batteries may no longer meet automotive range and performance requirements, but many of their electrodes can still function.

The Cornell method is not suited to every battery. It works best when performance loss is caused mainly by SEI buildup. Batteries with lithium loss, cracked particles, structural damage or mechanical failures would still need conventional recycling.

There are also practical hurdles. The electrodes must be removed from the battery, processed, and washed before they can be reused. Researchers tried injecting the DMI solution directly into intact cells, but the results were poor. That means a commercial version of the process would require battery disassembly and careful handling of the electrode materials.

Cornell estimates that DEER-recycled cells could cost $15.25 per kilogram, compared with $26.31 per kilogram for recycling through pyrometallurgy or hydrometallurgy. The estimate does not include recovering the DMI solution, which accounts for about 63% of the process cost. Reusing that chemical could improve the economics if the method moves beyond laboratory testing.

The research comes as battery recycling companies and researchers look for ways to handle a growing supply of aging EV batteries. While many of these methods remain focused on recovering metals, DEER offers a different option. Instead of immediately reducing a used battery to its chemical ingredients, it tries to salvage the electrode before that step becomes necessary.

Super Valid Designs spent two months on a ski lift because mice had moved into the chicken coop and ordinary traps only opened a revolving door. Catch one batch and another arrives. Live relocation felt like a better bet than a pile of snap traps, so the plan became a pair of jobs in one machine: baited gondolas that work as traps, then a cable line that carries those cars 250 feet down a hill and dumps the passengers into brush without anyone walking the route every night.

Mice travel to the vehicles using a gravity feeder that drops them from the coop ceiling. Each automobile is 3D printed and contains a portion of clear acrylic for display. The bottom door is smartly constructed to open by itself, eliminating the need for any electronics to interfere. When a car reaches the bottom station, the floor collapses beneath them, and our little critters land on a mound of brush. The entire thing is popularly referred to as the Rodent Tram Express; whether you interpret this as a show of kindness or a sardonic joke about a peanut-butter-fueled ride to freedom is totally up to you.

Bambu Lab A1 3D Printer, Ready-to-Use FDM 3D Printer for Beginners and Kids

• High-Speed Precision: Experience unparalleled speed and precision with the Bambu Lab A1 3D Printer. With an impressive acceleration of 10,000 mm/s…
• Multi-Color Printing with AMS lite: Unlock your creativity with vibrant and multi-colored 3D prints. The Bambu Lab A1 3D printers make multi-color…
• Full-Auto Calibration: Say goodbye to manual calibration hassles. The A1 3D printer takes care of all the calibration processes automatically…

A lot of man-hours went into laying the cable work; over 500 feet of it cost around $100 and resulted in a loop that was only about 250 feet long. He utilized old bike rims for the pulley wheels, covered them in some old tire rubber to give the cable some grip, and then had to get a windshield wiper motor from an old family car to make it work. After the rope sagged slightly and the cars began to get trapped on the support rods, the towers were installed. After that, it was basically a lot of framework made from whatever scraps he had lying around and measuring? Measuring stayed optional. “Measure twice, cut once, or don’t measure at all” was the working rule, with the cars themselves getting the careful design time the lift never received.

Testing was a pain in the neck, literally, because he had to hike the hill several times until all the kinks were worked out. Cars would snag on objects, the acrylic windows messed up the cable, and we even had a tower shift on us. If that wasn’t enough, he had to cope with heat and wildfire smoke, which turned our treks into a second project, as well as a host of ticks that appeared in the brush we needed to remove for the cable. He started by running two cars at once, just in case a full load of cars pulled the cable straight off the pulley wheels, but after a few runs, he eventually got everything in order and the loop held together.

To be honest, findings are a little hard to come by, since he only managed to get a few visits in with video evidence by the time he posted it all online. Whether 250 feet is truly far enough is debatable; many people pointed out that mice have a nasty habit of simply walking home after a while, and 1/4 mile is probably about right for a commute rather than exile. Super Valid Designs is considering the lift as a temporary fix, and he isn’t thinking of it as a silver bullet answer just yet. Keeping the feed sealed and holes plugged remains the top priority; the Rodentram Express is more of a nice-to-have than a must-have, but hey, we now have a station in the coop and a line on the hill, and a few mice have had the dubious pleasure of taking public transportation without asking.
[Source]

In context: The FAA has long suffered from a shortage of air traffic controllers, and many air traffic control systems are badly outdated, making the sector an obvious target for AI-powered automation. Although no one currently believes AI can replace air traffic controllers, the technology could help different airlines and the FAA pool data to increase efficiency.

Federal aviation officials aim to launch a new AI-based air traffic control tool next week in Washington, DC, one of the US’s busiest air travel corridors. The debut marks the beginning of a gradual nationwide rollout.

Strategic Management of Airspace, Routes and Trajectories (SMART) is a cloud-based system that analyzes schedules, weather patterns, airport capacity, and other operational factors to predict traffic flow and potential conflicts. While airlines, air traffic controllers, and the FAA currently often operate on different data sources and schedules, SMART aims to provide them with a shared intelligence stream.

Air Space Intelligence, the AI firm that developed SMART after receiving a $875 million grant from the FAA, hopes it will reduce fuel use, speed recovery following bad weather and congestion, raise situational awareness, and create space for new air traffic. Along with another system called Flow Management Data & Services, ASI aims to roll out SMART over the next one to two years. The initiative is part of the Trump administration’s plan to update US air traffic control, which has struggled with staffing shortages and outdated technology for years.

A 2023 FAA risk assessment found that 37% of systems were unsustainable and 39% were potentially unsustainable. While plans to update some systems could take a decade or more, others – some over three decades old – were not scheduled to be modernized. ASI is promoting SMART as a faster way to update some of the FAA’s tools.

AI proponents also often cite the technology as a path to reducing the workloads of overburdened employees (though many claim it does the opposite), and air traffic controllers could be among those who benefit. While the FAA hopes to eventually employ 12,563 controllers (down from a prior target of 14,633), it currently employs only around 11,000. The shortage has contributed to flight delays and made one of the world’s most stressful jobs even more taxing. The FAA has even resorted to video games to boost recruitment, but the job’s mental requirements are steep.