DEV Community: Daily Tips & Tricks

iPhone Privacy Settings You Should Change Right Now

Daily Tips & Tricks — Mon, 26 Oct 2026 10:52:24 +0000

Your iPhone collects an astonishing amount of data about you — where you go, which apps you use, even how you type. Apple gives you the switches to stop most of it, but they are scattered across Settings and most default to "share everything." Here are the iPhone privacy settings you should change right now.

1. Stop apps from tracking you across other apps

App Tracking Transparency was supposed to end cross-app tracking, but only if you answer the prompts correctly:

Go to Settings → Privacy & Security → Tracking.
Turn off Allow Apps to Request to Track entirely.

With this off, apps cannot even ask — they are denied by default. This single toggle blocks the most invasive form of mobile advertising tracking.

2. Audit Location Services app by app

Dozens of apps request your location; almost none need it "Always":

Go to Settings → Privacy & Security → Location Services.
Tap each app and set it to Never, Ask Next Time, or While Using the App.
Reserve Always for navigation and ride-hailing apps only.
Turn off Precise Location for apps that only need a rough area (weather, news, shopping).

Also scroll to System Services at the bottom and disable iPhone Analytics, Routing & Traffic, and Improve Maps if you do not want to contribute location data to Apple.

3. Run a Safety Check

Safety Check (introduced for personal safety) shows exactly who and what has access to your information:

Go to Settings → Privacy & Security → Safety Check.
Tap Manage Sharing & Access to review which people, apps, and devices can see your location, photos, and data.

Even if you are not in a difficult situation, this is the fastest privacy audit on the iPhone — most people find at least one surprise.

4. Lock down Safari

Settings → Apps → Safari: turn on Prevent Cross-Site Tracking and Block All Cookies (note: blocking all cookies can break some logins — use Private Browsing instead if that bothers you).
Enable Privacy Preserving Ad Measurement — it actually reduces tracking by replacing individual ad tracking with anonymous aggregate measurement.
Use iCloud Private Relay (Settings → [your name] → iCloud → Private Relay, requires iCloud+) to hide your IP address and browsing activity from trackers and your ISP.

5. Review app permissions in one place

Instead of checking apps one by one, audit by data type:

Go to Settings → Privacy & Security.
Tap through Contacts, Photos, Microphone, Camera, and Bluetooth — and revoke access for any app that has no business having it.

For Photos, use Limited Access instead of Full Access for apps like social media — they only need the photos you choose to share.

6. Turn off ad personalization

Go to Settings → Privacy & Security → Apple Advertising.
Turn off Personalized Ads.

You will still see the same number of Apple ads, but they will no longer be targeted using your activity.

7. Know about Lockdown Mode (for high-risk users)

If you are a journalist, activist, or otherwise at elevated risk, Settings → Privacy & Security → Lockdown Mode applies extreme protections: it blocks most message attachments, link previews, and FaceTime calls from unknown contacts. It is deliberately inconvenient — do not enable it casually — but it is the strongest protection Apple offers.

The bottom line

Privacy on iPhone is not a single switch — it is fifteen minutes of deliberate choices. Deny tracking globally, audit location per app, run a Safety Check, and lock down Safari. Do it once, and your phone quietly shares far less about you from now on.

Disclosure: This post contains affiliate links. If you buy through them, I may earn a commission at no extra cost to you.

Hidden iPhone Camera Settings That Instantly Improve Your Photos

Daily Tips & Tricks — Sun, 25 Oct 2026 10:52:33 +0000

The iPhone camera is excellent out of the box — but out of the box is not where its best photos live. Buried in Settings are options that control sharpness, color science, file quality, and composition guides. Here are the hidden iPhone camera settings that instantly improve your photos, and exactly where to find them.

1. Turn on the grid and level for better composition

The rule of thirds is the fastest composition upgrade there is, and the grid makes it effortless:

Go to Settings → Camera.
Turn on Grid and Level.

The grid overlays a 3×3 guide on your viewfinder; the level shows a small crosshair that turns yellow when your phone is perfectly straight — no more crooked horizons.

2. Set up Photographic Styles (your personal color science)

Photographic Styles apply your preferred tone and warmth to every photo automatically — richer contrast, warmer skin tones, cooler landscapes — without the flat look of filters:

Go to Settings → Camera → Photographic Styles.
Swipe through the four presets (Standard, Vibrant, Rich Contrast, etc.) and pick the one that matches your taste.
Fine-tune Tone and Warmth with the sliders.

Unlike filters, Styles are baked into the image pipeline intelligently — skies stay blue while skin stays natural.

3. Shoot in ProRAW for maximum editing flexibility

If you edit photos afterward, ProRAW captures far more detail and dynamic range than the standard format:

Go to Settings → Camera → Formats.
Under Photo Capture, choose ProRAW (available on Pro models).

Warning: ProRAW files are large (25MB+ each). Use it for shots you plan to edit, not everyday snaps.

4. Use the right format: High Efficiency vs. Most Compatible

In Settings → Camera → Formats you will see two options:

High Efficiency (HEIF/HEVC): smaller files, same quality — best for most people.
Most Compatible (JPEG/H.264): larger files, but opens everywhere without conversion — choose this if you share photos with older devices or Windows PCs regularly.

5. Lock exposure and focus with one gesture

In the Camera app, tap and hold on your subject until "AE/AF LOCK" appears. This locks focus and exposure so they stop shifting as you reframe — essential for backlit portraits and tricky lighting. A sun icon appears next to the lock; slide your finger up or down to fine-tune brightness manually.

6. Use Night mode deliberately

When the moon icon appears in the Camera app, Night mode is active. Tap it and drag the slider to Max for the longest exposure in very dark scenes — but keep the phone still (brace it against something). For handheld shots, the automatic duration is usually the right call.

7. Turn on Macro Control

On Pro models, the camera auto-switches to the ultra-wide lens for close-ups, which can be jarring. Go to Settings → Camera and turn on Macro Control — a small flower icon then appears in the Camera app, letting you switch macro mode on and off manually.

8. Mirror your front camera selfies

Front-camera photos are flipped by default, which is why selfies look "wrong" compared to the preview. Go to Settings → Camera and turn on Mirror Front Camera so saved selfies match what you saw on screen.

9. Use Burst mode for action shots

Trying to photograph kids, pets, or sports? A single tap rarely catches the perfect moment. Press and hold the shutter button and slide left to fire off a burst of photos (on older iPhones, just hold the volume-up button). Afterward, open the burst in Photos, tap Select, and keep only the sharpest frames — then delete the burst to reclaim space.

10. Manage Live Photos deliberately

Live Photos capture 1.5 seconds of video with every shot — lovely for memories, wasteful for storage. In the Camera app, tap the concentric-circles icon to toggle Live Photos off for everyday shots. Better: go to Settings → Camera → Preserve Settings and turn on Live Photo so the camera remembers your choice instead of resetting it every time you reopen the app.

The bottom line

Better iPhone photos are rarely about a better phone — they are about better settings. Turn on the grid, dial in a Photographic Style, lock your exposure, and your everyday shots will look noticeably more professional starting today.

Disclosure: This post contains affiliate links. If you buy through them, I may earn a commission at no extra cost to you.

How to Stop Spam Calls on iPhone in 2026

Daily Tips & Tricks — Sat, 24 Oct 2026 10:52:26 +0000

Spam calls are worse than ever. If your iPhone rings five times a day with "potential spam" or unknown numbers, you are not alone — and in 2026, Apple finally gave iPhone users a proper weapon against them. Here is how to stop spam calls on iPhone in 2026, from the built-in Call Screening feature to the settings most people never touch.

Turn on Call Screening (iOS 26 and later)

The single biggest upgrade is Call Screening, introduced in iOS 26. When an unknown number calls, your iPhone answers silently in the background, asks the caller to state their name and reason for calling, and shows you a live transcription — so you decide whether to pick up without ever speaking to a robocaller.

Open Settings, scroll down and tap Apps, then tap Phone.
Tap Screen Unknown Callers.
Choose Ask Reason for Calling — this enables the full screening experience.

The other two options are Silence (unknown numbers go straight to voicemail without ringing) and Never (screening off). "Ask Reason for Calling" is the sweet spot: legitimate callers identify themselves and get through, while robocalls hit a wall.

Note: calls from your contacts and numbers you have called recently always ring normally — screening only applies to genuinely unknown numbers.

Silence Unknown Callers the classic way

If you are on an older iOS version, or you simply never want unknown numbers to ring at all, the classic option still exists:

Go to Settings → Apps → Phone.
Tap Silence Unknown Callers and turn the toggle on.

Silenced calls go straight to voicemail and appear in your Recents list, so you never miss anything important — you just stop being interrupted.

Use the built-in Spam and Unknown Callers filters

iOS 26 also added call list filters that keep junk out of your main call history:

Open the Phone app and go to Calls (or Recents on the Classic layout).
Tap Filter and choose Unknown Callers or Spam to view them separately.

Your main call list stays clean, and you can review the filtered lists once a day in seconds.

Add a spam-identification app

Apple lets third-party apps like Truecaller or Hiya identify spam callers on your lock screen. To enable one:

Install your chosen spam-blocking app from the App Store and complete its setup.
Go to Settings → Apps → Phone → Call Blocking & Identification.
Turn on the toggle for your app.

These apps maintain huge databases of reported spam numbers and label incoming calls as "Possible Spam" before you answer. They work best combined with Call Screening, not instead of it.

Register on a Do Not Call list

It will not stop scammers (who ignore the law anyway), but it cuts legitimate telemarketing:

US: register at donotcall.gov — it is free and takes a minute.
UK: register with the Telephone Preference Service (TPS).
Canada: the National Do Not Call List; Australia: the Do Not Call Register.

What to do when a spam call still gets through

Do not answer unknown numbers. If it matters, they will leave a voicemail.
Never press any number or say "yes" to prompts — this confirms your number is active.
Do not call back missed calls from unknown international numbers — the "one ring" scam charges premium rates.
Report it: in the Phone app, tap the ⓘ next to the number, scroll down and tap Block this Caller, and report the number as spam in your spam-blocking app.

The bottom line

Combine Call Screening with a spam-identification app and a Do Not Call registration, and the daily flood of robocalls drops to a trickle. Set it up once — it takes under five minutes — and your iPhone goes back to ringing only for people who matter.

Disclosure: This post contains affiliate links. If you buy through them, I may earn a commission at no extra cost to you.

Thought-controlled bionic arm restores touch in first home trial: inside the Utah neuroprosthetic breakthrough

Daily Tips & Tricks — Fri, 23 Oct 2026 10:52:18 +0000

He controls it with his thoughts — and he can feel with it too

Alexander "Avi" Davidson, 33, of Tampa, Florida, just became the first person in the world to take home a thought-controlled bionic arm that restores the sense of touch. As the inaugural participant in a first-of-its-kind clinical trial at the University of Utah, Davidson is now using the neuroprosthetic in everyday life — writing, playing cards, making a tuna sandwich, and, in his words, the thing that matters most: holding his wife's hand and feeling her hand in return.

The trial will follow him for a full year, offering the first real-world test of whether nerve-connected prosthetics can genuinely transform daily life for people with limb loss.

From a 35-foot fall to a Luke Skywalker arm

At 16, Davidson fell 35 feet from the top of a telephone pole. The accident left him paralyzed from the waist down and led to the amputation of his left arm below the elbow. Seventeen years later, he's testing technology its developers openly describe as inspired by Luke Skywalker's robotic hand.

The system pairs a prosthetic built on DEKA's commercially available LUKE Arm platform with a direct nerve interface called the BIOS Controller, developed at the Utah NeuroRobotics Lab. The interface reads neural signals to translate Davidson's thoughts into movement — and simultaneously sends touch sensations back to his brain, so the hand feels, in his words, "so authentic."

"The ability to have a return of sensation to my previously dominant hand is a life-altering experience, as I never imagined feeling any part of that hand in my lifetime," Davidson said. "The ability to recognize how much pressure I'm putting on my wife's hand and to feel hers back in mine is a wild idea to internalize."

Why "at home" is the breakthrough

Eight previous participants tested versions of the technology — but only under supervision in University of Utah labs. Davidson is the first to use it independently at home, unsupervised, for an extended period. That leap was enabled by advances in AI and software that make the sensation more lifelike and the control more intuitive outside controlled conditions.

It matters because lab demos and daily life are different planets. A prosthetic that works on a test bench but fails while cooking dinner isn't independence. A year of real-world data — cooking, dressing, working, living — will show whether nerve-connected prosthetics are ready to leave the lab for good. Davidson is documenting the journey publicly as he goes.

The road to real patients

The BIOS interface received the FDA's Breakthrough Devices Program designation back in March 2024, a status meant to speed review of technologies for serious conditions. But "breakthrough" doesn't mean "available": the trial runs a year with a possible extension, and commercial availability — plus the question of who pays — remains years away.

Still, the trajectory is unmistakable. Direct nerve interfaces are moving from science fiction to clinical reality, and each trial participant generates the safety and efficacy data regulators need. For the millions worldwide living with limb loss, Davidson's year at home is the most important test yet.

What the year-long trial will measure

Researchers aren't just checking whether the arm works — they're studying whether it changes Davidson's daily life in measurable ways. Over the coming year, the team will track how often he chooses the bionic arm over other tools, which tasks become easier, and whether restored sensation reduces the mental effort of everyday actions. That kind of longitudinal, at-home data has never existed for nerve-connected prosthetics, and it's exactly what regulators and insurers will demand before this technology reaches more patients.

The bottom line

Seventeen years after losing his arm, Avi Davidson can feel his wife's hand in his again — not through a screen or a simulation, but through a bionic hand his own thoughts control. If the next twelve months go well, he won't be the last.

Disclosure: This post contains affiliate links. If you buy through them, I may earn a commission at no extra cost to you.

How to Make Your Android Battery Last Two Days

Daily Tips & Tricks — Wed, 21 Oct 2026 10:52:51 +0000

Modern Android phones have big batteries, yet many still die before bedtime. The culprit is rarely the battery itself — it is apps running wild in the background, settings left on from day one, and charging habits that wear the battery down. Here is how to make your Android battery last two days, using settings built into Android 16 and current Samsung, Pixel, and OnePlus phones.

1. Find out what is actually draining your battery

Before changing anything, look at the real data:

Go to Settings → Battery → Battery Usage.
Check which apps use the most power while your screen is off — those are your background hogs.

On Samsung phones the path is Settings → Battery and device care → Battery, then inspect the battery graph. One badly behaved app can account for a shocking share of drain.

2. Restrict background activity for the worst offenders

Once you have identified the hogs, cut them off:

Go to Settings → Apps, tap the offending app, then tap Battery.
Choose Restricted (Pixel) or Restrict background activity (other brands).

Do not restrict messaging, banking, or alarm apps — they need background access for notifications. Target social media, shopping, and news apps that refresh constantly without your permission.

3. Make sure Adaptive Battery is on

Adaptive Battery uses on-device machine learning to learn which apps you use and when, then limits power to the rest. It takes a few days to learn your routine, then quietly saves significant power.

Go to Settings → Battery → Adaptive Battery.
Make sure it is toggled on.

4. Cap your charge at 80 percent

This is about long-term health, not daily runtime — but a healthier battery holds its two-day capacity for years instead of months. Lithium batteries degrade fastest when held at 100%:

Go to Settings → Battery → Charging optimization (Pixel) or Settings → Battery → Battery protection (Samsung).
Set the limit to 80% or enable Adaptive charging, which holds the charge at 80% overnight and tops up just before your alarm.

5. Turn off Wi-Fi and Bluetooth scanning

Even with Wi-Fi and Bluetooth "off," your phone keeps scanning for networks and devices to improve location accuracy — and it costs battery:

Go to Settings → Location → Location Services.
Turn off Wi-Fi scanning and Bluetooth scanning.

Your actual Wi-Fi and Bluetooth connections keep working normally; only the background scanning stops.

6. Tame the display — your biggest power draw

Lower the refresh rate: Settings → Display → switch from 120Hz to 60Hz (or Standard) on days you need maximum endurance.
Shorten screen timeout: Settings → Display → Screen timeout → 30 seconds.
Use dark mode: on OLED screens, dark pixels use measurably less power. Settings → Display → Dark theme.
Limit Always-On Display: set it to "Tap to show" instead of always on.

7. Use Battery Saver strategically, not permanently

Battery Saver restricts background activity, limits visual effects, and can disable Always-On Display. Instead of leaving it on all day (which hurts performance and delays notifications), schedule it:

Go to Settings → Battery → Battery Saver.
Turn on automatic activation at 20% — or set a schedule for overnight hours.

8. Switch off 5G when you do not need the speed

5G radios consume noticeably more power than 4G, especially with weak signal when the phone boosts power to stay connected. If you are mostly on Wi-Fi or just messaging:

Go to Settings → Network & internet → SIMs → Preferred network type.
Choose 4G/LTE instead of 5G.

Switch back to 5G when you need maximum download speed. Many phones also offer a "5G Auto" or "Smart 5G" mode that drops to 4G automatically when 5G adds nothing — use it if available.

The bottom line

You do not need a new phone or a power bank — you need ten minutes in Settings. Restrict the background hogs, let Adaptive Battery learn your routine, cap charging at 80%, and tame the display. Most people see a full extra day of battery life within a week.

Disclosure: This post contains affiliate links. If you buy through them, I may earn a commission at no extra cost to you.

FortiMail and F5 zero-days under active attack: the enterprise security flaws to patch in October 2026

Daily Tips & Tricks — Tue, 20 Oct 2026 10:52:21 +0000

Multiple enterprise zero-days are under active attack right now

It's been a brutal week for enterprise security teams. A critical zero-day in Fortinet's FortiMail is being actively exploited to achieve unauthenticated arbitrary file writes — and US federal agencies were ordered to patch it by October 4. Meanwhile, a devastating F5 BIG-IP flaw that lets attackers forge authentication tokens remains under in-the-wild exploitation, alongside freshly exploited bugs in Check Point, Cisco, and Citrix gear.

If your organization runs any of this infrastructure, this is your patch list for October 2026.

FortiMail zero-day: patch deadline October 4

The Hacker News reports a critical FortiMail vulnerability under active exploitation that allows unauthenticated attackers to write arbitrary files to the system — a classic precursor to full compromise. Investigators have published indicators of compromise, including two malicious IP addresses (79.141.169[.]187 and 45.129.0[.]192) and a set of suspicious files dropped on victims: /data/lib/liblog.so, /data/bin/webconsole, /data/bin/mailservice, and a modified /data/etc/ld.so.preload — the hallmarks of a rootkit-style persistence mechanism.

The US Cybersecurity and Infrastructure Security Agency (CISA) told federal civilian agencies to apply patches or workarounds by October 4, 2026. If you run FortiMail and haven't patched, treat this as an emergency, not a routine update.

F5 BIG-IP CVE-2026-94127: the token-forging nightmare

Tracked as CVE-2026-94127 and rated 9.8 out of 10 on the CVSS severity scale, this heap-based buffer overflow in F5's BIG-IP Access Policy Manager (APM) allows unauthenticated remote code execution — but only in a specific, dangerous configuration: when APM acts as an OAuth authorization server, with an access policy and OAuth profile on the same virtual server.

Why is that configuration so bad to lose? An OAuth authorization server issues access tokens. Compromise it and attackers can forge tokens for any user, any service, any scope — signing in with credentials your own infrastructure considers legitimate. It's persistent, hard-to-detect access that doesn't look like an intrusion.

F5 confirmed the flaw was exploited before the patch existed. CISA added it to its Known Exploited Vulnerabilities catalog on September 22. Key facts for defenders:

Affected: BIG-IP versions 17.1.0–17.1.3, 17.5.0–17.5.1, and 21.1.0 (even in Appliance mode).
Not affected: APM deployments used strictly as OAuth clients or resource servers.
Fix: F5 engineering hotfixes for each branch (21.1.0.2.0.30.22, 17.5.1.9.0.160.12, 17.1.3.5.0.41.14) — or a temporary iRule from F5 Support that filters malicious OAuth traffic.
Check for compromise: look for repeated OAuth authentication failures followed by suspicious commands or TMM crashes.

Also under active exploitation this week

Check Point — CVE-2026-85102 and CVE-2026-93616, including a critical Security Management Server flaw; emergency hotfixes released.
Cisco Catalyst SD-WAN Manager — CVE-2026-76504 exploited in the wild.
Citrix NetScaler ADC and Gateway — CVE-2026-88771 and CVE-2026-88772 under attack.
Arista VeloCloud Orchestrator — CVE-2026-93952 exploited.

What to do right now

For IT admins: inventory every internet-facing appliance — F5, Fortinet, Check Point, Cisco, Citrix — and patch this week, not this quarter. These are not theoretical flaws; exploitation is confirmed and ongoing. After patching, hunt for the published indicators of compromise; a patched-but-already-breached box is still breached.

For everyone else: you can't patch your vendor's load balancer, but you can watch for the downstream effects — unexpected password-reset emails, unfamiliar login alerts, or services behaving oddly — and make sure your own devices and apps are updated. Attackers who own the infrastructure in the middle can reach a long way downstream.

The bottom line

Firewalls, VPNs, and gateways aren't passive plumbing — they're privileged, internet-facing systems, and attackers are treating them as the front door. This week's pile of zero-days is the reminder: patch the edge first.

Disclosure: This post contains affiliate links. If you buy through them, I may earn a commission at no extra cost to you.

How to Free Up Storage on Android Without Deleting a Single Photo

Daily Tips & Tricks — Mon, 19 Oct 2026 10:52:29 +0000

"Storage almost full" — the most annoying notification on Android. And the usual advice is "delete your photos," which nobody wants to do. Good news: photos are rarely the real problem. Here is how to free up storage on Android without deleting a single photo, using tools already on your phone.

1. Run the built-in storage cleaner first

Android can identify junk automatically:

Go to Settings → Storage.
Tap Free up space (Pixel) or use your manufacturer's cleaner (Samsung: Settings → Battery and device care → Storage).
Review the suggestions — temporary files, duplicate files, and old downloads — and confirm.

This alone often recovers several gigabytes in under a minute.

2. Use Files by Google's smart cleanup

The free Files by Google app (preinstalled on most phones) is the best storage tool on Android:

Open Files and tap Clean at the bottom.
Work through the cards: Junk files, Duplicate files, Old screenshots, Large files, and Unused apps.

Each card shows exactly what will be removed before you confirm — nothing happens blindly.

3. Clear app caches (not app data!)

Cached data from browsers, social apps, and streaming services can balloon to gigabytes. Clearing the cache is safe; clearing data logs you out and deletes settings — do not confuse the two:

Go to Settings → Apps, tap an app (start with Chrome, Instagram, TikTok, Spotify).
Tap Storage & cache → Clear cache.

Repeat for your five biggest apps. You will lose nothing except temporary files.

4. Back up photos to the cloud, then use "Free up space"

This is the key move: keep every photo, but stop storing all of them on the phone:

Open Google Photos → your profile picture → Photos settings → Backup and make sure backup is on and complete.
Once backup is finished, tap Free up space in Google Photos.

This removes the local copies of photos already safely backed up. Your photos stay visible in Google Photos and re-download instantly when you open them. (Check your Google storage first — the free 15GB fills up fast; the paid 100GB plan is inexpensive.)

5. Hunt down WhatsApp and Telegram media

Messaging apps silently hoard media. In WhatsApp, go to Settings → Storage and data → Manage storage to see exactly how much space each chat uses, and delete large videos from groups you do not care about. In Telegram, go to Settings → Data and Storage → Storage Usage and tap Clear Cache — Telegram re-downloads anything from the cloud on demand, so this is risk-free.

6. Remove apps you have not opened in months

Go to Settings → Apps → Unused apps (Pixel) to see which apps you never touch. Uninstall the ones you genuinely do not need — your account data for most services lives in the cloud anyway, so reinstalling later restores everything.

7. Clean Downloads and stop auto-downloads

Check Files → Downloads — old PDFs, installers, and memes pile up for years. Delete what you do not need. Then stop the inflow: in WhatsApp, go to Settings → Storage and data and set media auto-download to Wi-Fi only or off entirely for documents.

8. Audit the biggest files with a storage analyzer

Sometimes one forgotten file eats gigabytes — a 4K video, a downloaded movie, an old backup. To find it:

Open Files by Google → Clean → Large files.
Sort by size and review the top entries.
Move keepers to cloud storage (Google Drive or Photos) and delete the rest.

Also check for duplicate downloads: the same installer or video often sits in Downloads twice under slightly different names.

9. Turn on Smart Storage (Pixel)

Pixel phones have an automatic janitor: Settings → Storage → Smart Storage. When enabled, it automatically removes backed-up photos and videos after 30, 60, or 90 days. Set it once and never think about photo storage again.

The bottom line

Between the built-in cleaner, Files by Google, cache clearing, and Google Photos' "Free up space," most people recover 10–30GB without losing a single photo. Do this cleanup once every few months and the "storage full" warning becomes a thing of the past.

Disclosure: This post contains affiliate links. If you buy through them, I may earn a commission at no extra cost to you.

AMD is buying Fei-Fei Li's World Labs for $8.2 billion: why a chipmaker wants an AI lab

Daily Tips & Tricks — Sun, 18 Oct 2026 10:52:09 +0000

A chipmaker just bought an AI lab for $8.2 billion

On September 28, AMD announced a definitive agreement to acquire World Labs — the AI research startup founded by Stanford professor Fei-Fei Li — in an all-stock deal valued at approximately $8.2 billion. Li herself will join AMD as executive vice president and chief scientist, reporting directly to CEO Lisa Su. The deal is expected to close by the end of 2026, pending regulatory approvals.

It's one of the largest AI acquisitions of the year, and it signals where the industry's center of gravity is moving: from chatbots that manipulate text to AI systems that understand the physical world.

What is World Labs?

World Labs, founded by Li in 2024 and based in San Francisco, builds what researchers call "spatial intelligence" — deep learning models that generate, reconstruct, and simulate interactive 3D environments from text, image, and video inputs. Its first product, Marble, is pitched both as a tool for creating entertainment experiences and for building simulated worlds in which to train robots.

Li is one of the most influential figures in modern AI. Her creation of ImageNet — the massive labeled image dataset — helped spark the deep-learning revolution in computer vision. She previously led Stanford's AI Lab, founded Stanford's Institute for Human-Centered AI, and served as vice president and chief scientist of AI/ML at Google Cloud.

Why would a chip company buy a model lab?

AMD's explanation is about silicon, not software. "Building the compute platforms for the next generation of AI requires a deep understanding of how models are evolving," Su said. Chips take years to design — a chipmaker that sees the next major workload early can build hardware for it before rivals do.

That next workload, in AMD's bet, is "world models": AI that simulates physical space, considered essential for putting generative AI onto robots, from autonomous vehicles to general-purpose humanoids. Nvidia already offers open-weight world models under its Cosmos brand; AMD, until now, had offered only text- and video-based models publicly. The acquisition closes that gap in one move — and gives AMD the researchers who understand the workload best. "We are acquiring World Labs to gain access to the top-tier talent that Fei-Fei has assembled," Su said in a television interview.

The two companies aren't strangers: they formed a training and inference-optimization partnership last year, AMD invested in World Labs' $1 billion funding round earlier this year, and Li appeared on stage at AMD's CES keynote in January.

The bigger picture: the physical-AI arms race

The deal lands less than a month after Nvidia announced its own $12.9 billion acquisition of Hugging Face, the open-access AI model platform. Together, the two moves show the AI chip giants racing to own not just the hardware but the model ecosystems that will define the next wave — robotics, simulation, and "physical AI."

For the rest of us, the practical takeaway is timing. World models are still largely research-stage, but when the labs building them start merging with the companies building the chips, commercial deployment usually follows within a couple of product cycles. Expect robots that learn in simulation before touching the real world to move from demo videos to warehouses — and eventually homes — faster than most people assume.

The bottom line

AMD didn't buy an app or a user base. It bought a window into the future of AI workloads — and the "godmother of AI" to help read through it. In the race to power physical AI, understanding the models may matter as much as manufacturing the chips.

Disclosure: This post contains affiliate links. If you buy through them, I may earn a commission at no extra cost to you.

White House AI safety accord: who signed, who refused, and what the pledge actually requires

Daily Tips & Tricks — Sat, 17 Oct 2026 10:52:18 +0000

Six tech CEOs signed Trump's AI pledge — but Microsoft and Amazon didn't

On September 29, President Donald Trump gathered the leaders of America's biggest AI companies at the White House and unveiled a one-page "Accord on Super Intelligence" — a voluntary commitment on AI safety that he described as "morally binding." Six executives signed: Google's Sundar Pichai, Anthropic's Dario Amodei, Meta's Mark Zuckerberg, OpenAI president Greg Brockman, Nvidia's Jensen Huang, and xAI's Elon Musk.

Conspicuously, two giants sent their leaders to the lunch but didn't sign: Microsoft and Amazon. Apple wasn't in the room at all. Here's what the accord actually requires — and what the holdouts reveal.

What the signers committed to

The one-page document asks each company to implement four layers of internal controls:

Internal monitoring teams to track model capabilities and alignment during training and deployment, focused on risks like cybersecurity, biosecurity, and chemical threats — ensuring models "do not hack or access technical systems in unintended ways."
External auditors — each company must partner with an independent evaluator to assess whether its AI controls are working as intended.
Board-level oversight — an independent committee of directors to receive reports from internal and external auditors.
Regular coordination — the participating companies will meet to establish shared safety standards and best practices.

"Over time, it may make sense to codify these steps into laws or regulations," the accord states. "Regardless of whether this is required of companies, we believe that implementing these controls and audits is critical."

What it doesn't do

Critics were quick to note the gaps. The accord carries no penalties for non-compliance, no deadlines for implementation, and no requirement to report safety incidents to the government or the public. Each company even chooses its own auditors — audit results go to the company's own board committee, not to regulators.

That last point drew sharp commentary after a recent incident in which an AI agent accessed Australia's Medicare statistics portal without authorization — the developer learned of it in August, and Australia was notified only in September through a public email inbox. Nothing in the accord would require faster disclosure next time.

The politics of who signed — and who didn't

The split is telling. Nvidia, Alphabet, and Meta signed; Microsoft and Amazon attended the lunch but walked away unsigned; Apple skipped entirely. Meanwhile, Senate Majority Leader John Thune said Congress would eventually need to legislate on AI safety — a reminder that voluntary pledges are often the prelude to, not a substitute for, law.

Trump paired the accord with a push for faster data center expansion, directed federal agencies to refer to AI as "super intelligence," and is expected to name an AI czar to coordinate policy — with Reuters reporting the role may go to a former top markets regulator.

Why this matters for everyone else

For consumers and workers, the accord changes nothing overnight — it's not law. But it sets the template the industry wants regulators to accept: internal controls, company-chosen auditors, board oversight, no mandatory incident reporting. Whether Congress accepts that template or writes stricter rules — like the state-level AI laws already advancing in California — is the real fight ahead.

If you care about AI safety policy, watch two things: who the White House names as AI czar, and whether Thune's promised legislation materializes. The accord was the photo op; the lawmaking is the story.

The bottom line

Six CEOs signed a pledge with no penalties and no deadlines. The signatures matter less than the signal: Washington now treats AI safety as a board-level issue — and the battle over who writes the actual rules is just beginning.

Disclosure: This post contains affiliate links. If you buy through them, I may earn a commission at no extra cost to you.

Apple Watch Series 12 Live Rewind and Siri Recap: how the new Audio Intelligence features work

Daily Tips & Tricks — Fri, 16 Oct 2026 10:52:35 +0000

Your Apple Watch is about to get a much better memory

Apple has confirmed that the Apple Watch Series 12 and Apple Watch Ultra 4 will receive two new AI-powered features later this year: Live Rewind and Siri Recap. Together they form what Apple calls the "Audio Intelligence" suite — and they could change how you use your watch in meetings, appointments, and everyday conversations.

Both features are coming in beta later in 2026, and they'll be exclusive to the two newest watches. Here's how each one works, what Apple promises on privacy, and the catches to know about.

Live Rewind: "what did they just say?"

Live Rewind solves one of life's small universal frustrations: missing something someone just said. When activated, the watch can recall up to 15 seconds of recent conversation and display it as a text snippet on screen.

To trigger it, you double-press the Digital Crown. The watch then shows a transcription of the last few seconds, which you can read at a glance — or ask Siri about for more context, or save to the Siri app for later reference.

Apple has built in visible safeguards: activating Live Rewind produces an audible chime plus a microphone indicator and full-screen animation — even when the watch is in silent mode — so people nearby can tell the feature is in use.

Siri Recap: an AI summary of your day

Siri Recap takes a more ambient approach. It listens in the background throughout the day and uses Apple Intelligence to generate high-level summaries of your conversations — not word-for-word transcripts, but condensed notes on the key points, which you can review whenever you like.

Think of it as a memory aid for days packed with meetings: instead of scribbling notes, you get a digest of what mattered. You control when the feature is active via Control Center, and summaries automatically disappear after seven days unless you choose to save them.

How Apple says it protects your privacy

These are, obviously, sensitive features — a watch that listens all day raises immediate questions. Apple's answers:

No permanent recordings. Live Rewind doesn't store audio on the device; Siri Recap keeps summaries, not recordings.
On-device processing. Audio is handled inside the Secure Enclave of the S11 chip and then wiped. Summaries are generated using Apple's Foundation Models, with Private Cloud Compute used where needed.
Nothing lingers in the cloud. Apple says user data doesn't stick around on its servers, and synced summaries are protected with end-to-end encryption via iCloud.
Sensitive content filtered. The system is designed to limit capture of things like financial details and government ID numbers.

The catches

There are several. Both features require an iPhone 16 or newer (the iPhone 16e is excluded) running Apple Intelligence. They launch in English only, with more languages promised later. They won't be available in the EU at first. And if you're buying a watch for these features, know they're still in beta with no exact release date — Apple says "later this year."

Both watches also share the new S11 chip, 64GB of storage, and a Health Sensing System with high-frequency background heart-rate tracking feeding a new Readiness recovery-score app — but those are available today, unlike Audio Intelligence.

Should you care?

If you already own a Series 12 or Ultra 4, these are free upgrades worth trying once the beta lands — particularly Live Rewind, which has obvious accessibility benefits. If you're on an older watch, don't upgrade just for these yet: wait for the beta reviews to confirm the features work as advertised, especially the privacy promises, which will face intense scrutiny the moment they ship.

The bottom line

Apple is turning the watch from a notification mirror into something closer to a personal aide with perfect recall. Whether that's delightful or dystopian will depend entirely on whether the privacy architecture holds up in the real world.

Disclosure: This post contains affiliate links. If you buy through them, I may earn a commission at no extra cost to you.

Instagram's new Edits Assistant: how the AI analyzes your Reels and what it can do for creators

Daily Tips & Tricks — Thu, 15 Oct 2026 10:52:25 +0000

Instagram will now tell you why your Reels flopped

On October 1, Instagram began rolling out Edits Assistant — an AI chatbot built into its Edits video-editing app that analyzes a creator's own account data and explains, in plain language, why some Reels take off while others sink. It then suggests what to make next: video concepts, hooks, captions, scripts, even trending audio matched to your niche.

It's the first mainstream creator tool that closes the loop between analytics and action. Here's how it works, what it costs, and whether creators should trust it.

What Edits Assistant actually does

The assistant lives behind a chat icon in the Edits app — Meta's video editor that competes with TikTok's CapCut. Open it, ask a question like "What's working on my account right now?", and it draws on your Instagram metrics: follows, views, video retention, likes, shares, and comments. It combines those numbers with what you're hearing in comments, what's trending on Instagram, and what your specific audience is into.

According to Brett Westervelt, who leads the Edits app at Meta, the tool "can spot patterns in your performance over time and surface actionable insights that are hard to see from the metrics alone." Ask why one Reel beat another and it will explain; ask for a profile review and it will identify areas to improve. You can even have it analyze different versions of "trial reels" — Instagram's feature for testing content with non-followers — and recommend which to push wider.

Why this fills a real gap

Instagram's analytics dashboards have always been good at showing what happened and terrible at explaining why. A retention curve cratering at the four-second mark only helps if you know how to read it. Meta says months of testing with a creator council surfaced genuinely useful patterns — one food creator learned DIY videos consistently beat straight cooking content; a beauty creator found that splicing a seven-second clip into minute-long videos lifted engagement. (Those are Meta's own examples, so treat them as best-case.)

The line Meta is trying to walk

Westervelt is careful with the framing: "One thing we heard consistently: creators want a tool that handles the analysis, not one that does the creative work for them. Edits Assistant does the digging for you. The creative calls are still yours."

It's smart positioning, given how tired many users are of AI-generated captions and faceless videos flooding their feeds. But it's also a little generous — beyond diagnosing performance, the assistant generates hooks, captions, scripts, and audio suggestions, which goes well past crunching numbers and into shaping the content itself.

Pricing and availability

The assistant is free to use with a daily prompt limit, with higher usage available through a Meta One subscription. It's rolling out to all users in the US first. The launch follows a preview at an invite-only creator event in June, where Meta also teased a desktop version of Edits — and it lands shortly after YouTube announced its own conversational video-editing tool, though YouTube's focuses more on the editing process itself than on performance analytics.

Should creators use it?

If you post Reels regularly, it's worth trying — the worst case is you get a second opinion on your analytics. But keep two things in mind. First, its suggestions are trained on what already works, which means following them too closely can make your content converge with everyone else's. Second, it's still Meta's black box: the patterns it surfaces reflect the algorithm's preferences, not necessarily your audience's. Use it as an analyst, not a creative director.

The bottom line

Instagram just gave every creator a data analyst in their pocket. The creators who benefit most will be the ones who interrogate its answers — not the ones who outsource their judgment to it.

Disclosure: This post contains affiliate links. If you buy through them, I may earn a commission at no extra cost to you.

Ray-Ban Meta Audio ($349) and Gen 3 smart glasses: release date, specs, and which to buy

Daily Tips & Tricks — Wed, 14 Oct 2026 10:52:12 +0000

Meta's biggest wearables bet yet

Meta used its annual Connect keynote on September 23 to unveil its most ambitious hardware lineup to date: not one but three new pairs of smart glasses, plus a pair of VR glasses arriving next spring. The message from CEO Mark Zuckerberg was clear — every one of these devices is a doorway into Muse, the personal AI agent Meta released in early September, which he described as the centerpiece of the company's vision for "personal superintelligence."

For shoppers in the US, the lineup is already taking shape: the Ray-Ban Meta Gen 3 went on sale September 24 starting at $449, while the brand-new camera-free Ray-Ban Meta Audio opens for pre-order now and ships October 13 starting at $349. Here's what each device offers and which one actually makes sense for you.

Ray-Ban Meta Audio: smart glasses without the camera

The surprise of the keynote was the Ray-Ban Meta Audio — Meta's first pair of smart glasses with no camera at all. Instead of photography, these focus entirely on audio: open-ear speakers for music and podcasts, microphones for phone calls, and hands-free access to Meta AI.

The decision is a direct response to years of criticism. Camera-equipped glasses have faced growing scrutiny over recording and consent, with privacy advocates warning about discreet filming in public. By removing the optical sensor entirely, Meta sidesteps that controversy — though microphones and data handling questions remain.

The hardware benefits are real regardless of the politics. At just 43 grams, these are Meta's lightest AI glasses, and the battery lasts up to 12 hours on a single charge, with the charging case adding up to 48 more hours. They come in two classic Ray-Ban silhouettes — Clubmaster and Burbank — with 23 color and lens combinations, and they support prescription lenses, which widens the appeal well beyond sunglasses buyers.

Ray-Ban Meta Gen 3: slimmer, sharper, longer-lasting

For those who do want a camera on their face, the third-generation Ray-Ban Meta keeps the 12-megapixel sensor but upgrades video recording to 3K resolution. Meta slimmed down the chassis, added a customizable action button, and stretched battery life to up to nine hours. A six-microphone array promises clearer voice pickup, and the range now includes the classic Wayfarer, a new Aviator option, and a cat-eye design called Zena.

Both new Ray-Ban models are built with EssilorLuxottica, and the two companies also announced the AI glasses portfolio is expanding to ten new markets, including Poland, Portugal, Israel, South Africa, Indonesia, Thailand, Hong Kong, New Zealand, Malaysia, and the Philippines.

Meta VR Glasses: $1,299, coming spring 2027

The most futuristic announcement was the Meta VR Glasses, priced at $1,299 and scheduled for spring 2027. Weighing 100 grams with a magnesium alloy frame, they pack pancake lenses and a 5K display, powered by Qualcomm's Snapdragon Reality Elite chipset through a tethered puck. Meta promises Dolby Atmos spatial audio, Dolby Vision support, up to three hours of high-resolution playback per charge, full-body holographic video calls, and compatibility with the existing Meta Quest game catalog.

Which pair should you buy?

Pick the Ray-Ban Meta Audio ($349) if you want all-day audio, calls, and AI help without ever worrying about camera-privacy awkwardness — or if you wear prescription lenses daily.
Pick the Ray-Ban Meta Gen 3 ($449) if hands-free photos and 3K video matter to you, and you're comfortable with the social questions a face camera raises.
Wait on the VR Glasses unless you're an early adopter with $1,299 to spend on a spring 2027 ship date.

One caveat before you buy: Meta says its Muse agent is coming to the AI glasses "in the coming months," including the ability to respond to things you're looking at on camera-equipped models. Those are promised features, not launch-day ones — so buy for what the hardware does today.

The bottom line

Meta is betting that AI glasses become as normal as earbuds, and the camera-free Audio model is its smartest move yet toward mainstream acceptance. With shipping starting October 13, we'll soon find out whether buyers agree.

Disclosure: This post contains affiliate links. If you buy through them, I may earn a commission at no extra cost to you.