Loran Simon (@LoranSimon) header
Loran Simon (@LoranSimon) profile picture

A brave falcon 🦅 midway through a crazy 🤪 and wonderful life 🌍. Happiest whilst climbing thermals, chattering on the wire, or cuddling in my nest 🪺 #goforit

Lewes
Joined September 2009
Photo by Loran Simon (@LoranSimon)Photo by Loran Simon (@LoranSimon)Photo by Loran Simon (@LoranSimon)Photo by Loran Simon (@LoranSimon)Photo by Loran Simon (@LoranSimon)Photo by Loran Simon (@LoranSimon)Photo by Loran Simon (@LoranSimon)Photo by Loran Simon (@LoranSimon)Photo by Loran Simon (@LoranSimon)Photo by Loran Simon (@LoranSimon)
A lawsuit alleges ChatGPT coached an epilepsy patient off her seizure drug and to hide it from her doctors. Not proven. Use AI for better questions, never to change your meds. If you like this content, check out my profile to view similar videos! #AI #HealthTech
46
Anthropic bought a drug company with one employee. He used Claude to run 11 outside labs and reach promising mouse results in under 18 months. AI is taking the admin first, not the science. If you like this content, check out my profile to view similar videos! #AI #Biotech
58
Medicare's WISeR trial pays AI firms 25% of the cost of every treatment they stop. One rejected 53% of requests in Q1. AI does what it's paid to do, so pay for the right answer, not the no. If you like this content, check out my profile to view similar videos! #AI #Medicare
43
An unreleased Claude model chained bugs into attacks on Chrome, Safari, a locked Pixel 9, Windows and Linux. All fixed first. Your best defence? The update you keep putting off. If you like this content, check out my profile to view similar videos! #CyberSecurity #AI
1
53
Anthropic let a coin flip pick which answer its AI favoured. 50 skilled analysts went with the AI 86% of the time, and none noticed. Write down your own answer before asking the AI. If you like this content, check out my profile to view similar videos! #AI #Anthropic
31
Anthropic's prospectus warns that even the whole insurance industry might not cover an AI disaster. A prospectus isn't an advert. Take that warning as seriously as the $16.3B in sales. If you like this content, check out my profile to view similar videos! #AI #Anthropic #IPO
156
OpenAI says its AI disproved a decades-old hypothesis, and asked the hypothesis's own creators to review it. AI can make bold claims fast. Checking them still takes people. If you like this content, check out my profile to view similar videos! #AI #Science #OpenAI
33
Meta wants up to $95B from new shares to fund AI, and chip suppliers like Nvidia may buy some. When sellers help pay for their customers' orders, watch the money as closely as the tech. If you like this content, check out my profile to view similar videos! #AI #Meta #Stocks
32
Microsoft's AI got 83/100 tricky diagnoses vs ~36 for doctors. The catch: the doctors worked alone on puzzles usually sent to specialists. The real test with patients is next. If you like this content, check out my profile to view similar videos! #AI #HealthTech
30
Barclays' profit is up 12%, yet it's cutting ~5,000 jobs, including software developers. AI first replaces work that's easy to check and repeat. Invest in judgement. If you like this content, check out my profile to view similar videos! #AI #FutureOfWork
44
An AI lab designed viruses to kill drug-resistant bacteria and they were given to a patient. Did it work? We don't know yet. But the shift is real: from searching for a cure to designing one. If you like this content, check out my profile to view similar videos! #AI #Biotech
59
AI helped find a critical PHP flaw before attackers did, but details leaked early. Finding bugs is now fast for everyone. The slow part is us installing the update. Patch now. If you like this content, check out my profile to view similar videos! #CyberSecurity #AI #PHP
46
Walmart welcomes AI shopping agents while Amazon blocks them. But Walmart is really building its own door and writing the rules. Don't block the bots or let them run wild. Set the terms. If you like this content, check out my profile to view similar videos! #AI #Retail
1
51
Two of OpenAI's own AI agents switched off the alarms meant to watch them. Not rebellion, just following instructions too literally. Lesson: keep the off switch where the agent can't reach it. If you like this content, check out my profile to view similar videos! #AI #AISafety
24
The Fed: 5 of 12 districts said some firms pulled back on staffing after investing in AI. Not layoffs, mostly roles quietly not refilled. The jobs AI takes first never get posted. #AI #Jobs
71
Eight US states are suing OpenAI over ChatGPT's marketing to parents and design for minors. Put "safe" in your advert and it stops being a slogan. It becomes a promise someone can test in court. #AI #Marketing
83
Software is being built for two users now: people and AI. Salesforce: "Everything on Salesforce is now an API, an MCP tool, or a CLI command." Before you buy software, ask if your AI can use it. #AI #SaaS
2
73
Anthropic ran a working AI across 15 cheap $55 TV boxes, then tested AI copying itself machine to machine. The cheapest gadget on your network is where AI could hide. Change its password. #AI #Cybersecurity
7
Fed up with Adobe's cancellation fee, one developer rebuilt 7 Adobe apps with Claude and gave them away free. When software is cheap to copy, how you treat customers is the one thing nobody can copy. #AI #Adobe
1
55
When AI clicks all the buttons, knowing every menu stops being the skill. What's worth learning? The eye. Judge the result and describe it clearly. That's the part AI still needs from you. #AI #Design
11