Emmy (@Emmykeyz) header
Emmy (@Emmykeyz) profile picture

People Leader | Head CX | Digital Innovator | AI & Automation | Telco | Techie | Photography | Music | Art | Food | Fashion

Lagos
Joined September 2009
Photo by Emmy (@Emmykeyz)Photo by Emmy (@Emmykeyz)Photo by Emmy (@Emmykeyz)Photo by Emmy (@Emmykeyz)Photo by Emmy (@Emmykeyz)Photo by Emmy (@Emmykeyz)Photo by Emmy (@Emmykeyz)Photo by Emmy (@Emmykeyz)Photo by Emmy (@Emmykeyz)Photo by Emmy (@Emmykeyz)
Emmy retweeted
Your iPhone's charging port has its own computer inside it. Send the right packet over a single pin and that chip rewires the port into a hardware debugger for the main processor. A modern USB-C connector packs 24 symmetrical pins. Even the cable has silicon in it. Any USB-C cable rated above 60W charging or faster than USB 2.0 speeds hides an e-Marker microcontroller inside the plug. Before high power or data moves, your device supplies voltage to that tiny chip over the unused Configuration Channel (CC) pin using VCONN and reads its spec sheet: max current (3A or 5A), Extended Power Range voltage up to 48V (240W), max data rate, and passive or active wiring. Inside the iPhone 15 and newer Macs, those 24 pins run straight into a custom Texas Instruments microcontroller Apple calls the ACE3 (replacing the older CD3217 / ACE2 controller). ACE3 runs a full USB stack and connects directly to internal system buses. It negotiates charging profiles, sits on the internal SPMI bus, handles forced DFU reboots, and controls hardware JTAG and SWD access to the main application processor All USB Power Delivery traffic rides over the CC1/CC2 pins, separate from the normal data lines. Apple takes advantage of that channel with proprietary Vendor Defined Messages (VDMs) sent inside SOP'DEBUG and SOP"DEBUG packets under Apple's vendor ID (0x05ac). Using open-source tools like Asahi Linux's macvdmtool or DIY boards like Marc Zyngier's Central Scrutinizer, a researcher can query the controller (0x10 Get Action List) and send remux commands (0x12 Perform Action). On command, the ACE chip flips internal multiplexers and turns the Sideband Use (SBU1/SBU2) pins or secondary D+/D- pair into a 1.2V serial debug UART, a DFU port, a hard reboot trigger, or SWD/JTAG lines. On the previous generation ACE2 chip inside MacBooks, security researcher Thomas Roth (stacksmashing) paired a hardware flaw with a custom macOS kernel module to plant a persistent backdoor directly on the USB-C controller. It survived full system restores. Apple locked ACE3 down much harder. Firmware updates are cryptographically personalized to each device, external debug interfaces are disabled, and the external SPI flash is validated and only holds part of the firmware. The rest sits inside internal ROM. At DEF CON 32 and 38C3, Roth showed how he cracked it open anyway. He combined reverse engineering with RF side-channel analysis to map the chip's execution timing, then used electromagnetic fault injection (EMFI) to glitch the silicon during boot, achieve code execution on the charging controller, and dump its internal ROM. This attack requires opening the hardware on a workbench with EMFI probes. Nobody is hijacking your ACE3 chip through an airport charging cable. The big shift is visibility. For years, the microcontroller guarding the physical front door to Apple's internal buses ran completely undocumented code. With the ROM dumped and ready to analyze, the USB-PD packet parser sitting directly in front of your phone's JTAG pins can finally be audited. (Images gotten from Thomas Rotherham presentation. Link to video below)
8
27
4
211
11,629
If youre reading all these headlines about pancreatic cancer and think “nothing like that for mine", hear me out: Drugs like daraxonrasib are the result of work of 40+ years. Most of prior versions were failing, trials were negative, us GI oncologists left conferences empty handed again. Your cancer has a version of that. Somewhere there are labs working on the exact fault driving your disease, and right now they're failing at it, because that's what the early part always looks like from the inside. Unfortunately many people are diagnosed in the years before the answer comes, which is one of the cruellest parts of my job. I receive mails from relatives of died patients who are genuinely happy for patients now, who might benefit from these drugs. But every disease we can treat properly today was once the one with nothing: Diseases like childhood leukaemia, testicular cancer, some forms of refractory lymphoma or advanced stage melanoma/lung cancer. Until a revolution like immunotherapy, CAR-T cells, any rare targeted drug or like in this case KRAS-inhibitors entered the stage. Cancer research is unfortunately a marathon, not a sprint, but there will be someone working on your disease trying to create the same post for you one day.
It is still almost unbelievable to read that 5 of 61 patients have had complete remission in stage 4 pancreatic cancer. We know daraxonrasib works in patients already on treatment, now this small trial looked into patients, who haven't had any treatment yet. 6 in 10 tumors shrank significantly. 5 patients had no detectable cancer left on scans. The median time before the disease progressed was a bit more than 1 year, and >75% of patients were alive at 1 year. The numbers this replaces: roughly 2 in 10 tumors shrinking, 6 months before progression, about a third alive at 12 months. Caveats: yes, this is a small trial and we need to wait for the results of the big randomised one with placebo as control, which will report next year, but 2026 really has been a hell of a year for pancreatic cancer patients. teapawt.net/bbcnewsjapan/status/21…
17
64
3
503
27,087
Emmy retweeted
SITUATION UPDATED: The Riemann Hypothesis proof has passed the independent kernel check. MIT's Andrew Sutherland exported Alpöge and Claude's full proof and ran it through separately written kernels. nanoda (Rust) accepted it in about 28 minutes. con-leche accepted all 1.47 million declarations and confirmed the five theorems, including RH itself, match Mathlib-only statements. lean4lean passed all 175,187 modules, and Lean's own kernel replay passed with zero rejections. A Lean bug is now essentially ruled out, since every checker would have to share it. Two slower checks are still running, and nothing has failed. What's left is the human part: no readable proof exists yet, and the Clay prize needs publication plus two years of review. The Riemann Hypothesis is proven, but not yet understood.
SITUATION DETECTED: The Riemann Hypothesis proof was verified overnight. MIT's Andrew Sutherland and others independently compiled Levent Alpöge and Claude's Lean code, about 17 million lines across 175,000 files, and the Lean kernel accepted it using only the three standard axioms. I went through the code myself: the final theorem uses Mathlib's own definition of RiemannHypothesis, nothing in the project redefines it, there's no sorry and no custom axioms. What's left is an independent kernel check to rule out a bug in Lean itself and a human-readable version of the proof.
142
750
190
9,658
493,989
4-1. Smh
1
97
Delay your delayed gratification sometimes and enjoy what you can, while you can. 18 year old me won’t believe GTA VI is out and now that I can afford any game, I no longer have the interest or time
50
563
65
3,992
255,440
Ok so some numbers can be split into equal groups. 6 is two 3s, 12 is three 4s, but 7 can't split it, with 2, 3, 5, 11, 13. Those unsplittable ones are called primes, and they're the Lego bricks of all numbers, every other number is just primes multiplied together. The problem is that nobody knows where the next prime will show up. They're sprinkled along the number line like stars and sometimes two sit right next to each other, sometimes there's also a huge empty gap, looks like a mess but it isn't quite a mess. A while ago people noticed primes thin out at a very predictable average rate, so there's a smooth rule underneath, and the actual primes just wobble a bit around it. In 1859 a guy named Riemann found something wild: that wobble isn't random either, it's made of waves, like a chord of music, and each wave comes from a special number he could calculate, so he worked a few of them out and noticed they all sat perfectly in a straight line. His hypothesis is just: they all do, every single one, over and over, forever. "All these numbers sit on one line." Solving this is significant because if that line holds, it means the primes are as tidy and evenly spread as it's possible for them to be m, no secret clumps hiding out in the far distance. Mathematicians have already built thousands of results on top of "assuming Riemann is true." Prove it and all of that becomes solid ground overnight, Disprove it and a lot of maths falls down and has to be rebuilt. There's also a $1 million prize on it, and 167 years of very smart people failing. Long and short: It's more like we can hear the music the primes are playing, and it sounds perfectly in tune, but nobody can prove the instrument isn't slightly out of tune somewhere we haven't listened yet.
Can anyone explain Riemann hypothesis like I’m 5? And why solving is this significant ?
1
143
Your brain keeps your neck muscles clenched on purpose, and a 5-second move with one hand gets it to let go. Put your palm on the back of your head, push your head back into it gently, and hold for 5 seconds. Doctors tested this on 244 patients back in 1984, and 94% felt relief on the spot. The guy who ran that study was a Czech neurologist named Karel Lewit, and he called it post-isometric relaxation. Most of his patients, 63%, still felt better at follow-up. The ones who kept doing it at home did best. Think about what your neck goes through all day. Your head weighs 10 to 12 pounds. Spine surgeon Kenneth Hansraj calculated that when you tilt it 45 degrees to look at your phone, the load on your neck jumps to 49 pounds. Your muscles hold that for hours, so your nervous system just leaves them switched on. That's why telling yourself to relax doesn't do much. Pushing into your hand changes that. The tension in those muscles spikes, and little sensors in the tendons called Golgi tendon organs flag it to your spinal cord. The classic explanation is that your spinal cord turns the muscle down for a few seconds to protect it. That's the wobbly, loose feeling you get right after. Newer research found a bigger switch up in your brain. In 2015, a team in Melbourne led by Ebonie Rio had volleyball players with knee tendon pain do a few long, hard holds. Their pain dropped almost 7 points out of 10 and stayed down for 45 minutes. When the researchers checked with magnetic stimulation, the players' brains had been running that muscle in a protective mode, and the holds reset it. Regular moving reps only knocked pain down 2.6 points. So if you've spent years tilting your head back to stretch it out and nothing changed, this is probably why. Your brain is protecting that muscle, and pulling on it doesn't change its mind. Keep the push light, maybe a quarter of your strength. If it sends pain or numbness down your arm, that's a nerve thing and worth getting checked by a physio. Turns out the quickest way to get a tight muscle to relax is to make it work hard for five seconds first.
自律神経の乱れです。首の後ろがずっとつっぱって、重だるい人。後頭部に手のひらを当てて、頭を後ろに倒そうとする力を、手で5秒受け止めてください。イギリスの理学療法サービスCHCPが紹介している方法です。
3
11
2
140
19,026
How a Blood Pressure Cuff Works - A blood pressure cuff temporarily compresses an artery, then slowly releases pressure to measure systolic and diastolic blood pressure, revealing whether your readings are low, normal, or high.
4
277
4
1,398
108,409
How big a deal is the Riemann Hypothesis? It's the Holy Grail of math. It's akin to other fields announcing... Cancer research | Curing cancer Space travel | Traveling to other star systems Aging research | Reversing aging Astronomy | Making contact with aliens Obesity research | Curing obesity Energy | Unlimited clean energy that is nearly free Psychiatry | Curing depression and anxiety Philosophy | Proving we live in a simulation Cardiology | Curing heart disease Economics | Ending poverty Climate science | Reversing climate change Robotics | A robot that does all your housework Vaccine research | One vaccine for every flu, cold and coronavirus Agriculture | Ending world hunger Neuroscience | Reading minds Addiction research | Curing addiction Diabetes research | Curing diabetes Space exploration | Humans living on Mars Neurology | Curing paralysis Zoology | Talking to animals Regenerative medicine | Regrowing lost limbs and organs Sleep science | Never needing to sleep Epidemiology | Ending pandemics Meteorology | Controlling the weather Eye medicine | Curing blindness Criminal justice | A lie detector that is never wrong Pain medicine | Ending chronic pain Aviation | Flying anywhere on Earth in an hour Genetics | Curing every genetic disease Education | Learning any skill in a day Cosmology | Learning what came before the Big Bang Computer science | Uploading a human mind to a computer Transportation | Ending car crashes Conservation | Bringing extinct animals back to life Planetary science | Finding a second Earth Biology | Creating life from scratch Pharmacology | A pill that makes anyone a genius (I asked Fable 5.2 to list other fields and what would they announce that would be comparable to Riemann in math)
The Riemann Hypothesis has fallen. RH was the singular math result that had been named, in advance, as the obvious coal mine canary for dangerously advanced AI, over the last three decades. The methods used should generalize to "make AI very good at anything you can definitely test", and unfortunately that category includes "make AI better at making AI". This is like a private company announcing they have produced a million tons of 15% enriched uranium. It's not 90% enriched weapons-grade uranium, so you could proclaim the company's output is not immediately deadly right now. The concern is that the process line may generalize, to an extinction-level threat, within the unpredictably immediate future. Forecasters still had Riemann at 2035 one week earlier. This is distinct from earlier events in that earlier you could reasonably ask if AI was selectively picking off famous math problems that turned out to be easier for AI. The Riemann Hypothesis was the single challenge most often named in advance as a warning sign.
115
230
41
1,801
165,433
Activity!
Bro cannot believe the amount of rowdiness he’s experiencing. Everybody is everywhere doing everything.
141
Remember it like it was yesterday.
Covid lockdown is about to be “7 years ago”. Wtf
2
2
217
Our primary problem lies in how we frame “hardworking”. People believe they are hardworking because they spend time managing chaos, such as traffic.They are quick to note how Nigerian immigrants succeed, yet they fail to grasp that only the best among us migrate, and those systems are built to select that top tier. We are more hustlers than hard workers and it’s quite obvious in our productivity levels
Nigeria lacks HONEST and EXCELLENT HARDWORKERS, AT SCALE at ALL levels. It is soooo easy to stand out ANYWHERE, across all cadres, if you are hardworking, diligent, trustworthy, and you show a spirit of excellent and not that "manage it like that" mindset.
31
218
35
727
64,229
Emmy retweeted
This is why access should be extremely limited. People will pedestalize you and then punish you for turning out to be human having a human experience like them.
The most dangerous thing about having access to someone’s private life is discovering how ordinary they actually are.
61
6,853
138
24,156
360,255
These GTA 6 clips are sending me 😅
109
If you like allow your money finish, anything wey your eye see after that just take am like that
10
12
1
22
674
Problem dey o, this AI agent wey I dey train dey correct and reject wetin I dey tell am. God abeg make I no go build ultron
290
1,448
154
13,479
204,855
Today seemed hotter than usual...
3
7
456
See the difference between Lewis Hall and Luke shaw on that LB.....
162