Father, Husband, Dog lover 💙💜❤️| Founder @tixperienceapp & https://xitter.cf/t.co/xToa2mPLzk
UK Decacorn status (loading)
Joined July 2011
- Tweets32.4K
- Following3.6K
- Followers4.4K
- Likes149K
Pinned Tweet
2017: Helped raise over $250 million USD for projects at ICO and TGE phases. Featured in NYT Best Seller 'Future Proofing You', from Jay Samit. Ran worlds largest Growth Hacking and Copywriting group, hosted london events and released Growth Hacking Bible - Ace The Game, follow up to best seller the "Secret Sauce" from Austin Alred and Vin Clancy.
2018: Helped launch 3 products that still get used in the space by thousands of users and 6 figures of wallets a month. Protocol isn't sexy but survives and lots of tech in the space is based on what we helped test, ideate and launch.
2019: Settled into a couple of roles to build out friends companies as a side hobby in Web3. whilst also pivoting to more advisorial roles for brands looking to educate their teams on erc721 and how collectibles can be utilised. Panelist, Speaker, Host.
2020: Blew up on Clubhouse and was a lead educator in rooms, moved to Twitter - but got pegged back by a select few gatekeepers that discarded a few of us once they learnt what they needed and passed off the knowledge as their own. People forget you when you don't attend all the glitzy events that those with lots of followers do...
2021: Brought forth the worlds first virtual punk-emo girl-band. Launched digital collectible cards and wearables within metaverses. Presented and produced acts for Decentralands Metaverse Music Festival. A World first. Launch partner on Elynixr, powered by the same tech as Pokemon Go!
2022: Got stuck in VC turmoil ying-yang land where no one would wish to lead any rounds but all wished to enter them... Brought 2nd headliners to DCL MMF again and our own stage. Nominated for dozens of top ai, nft, tech awards across the year.
2023: Got burned and lost over $1million USD on a project that the founder rugged his whole staff/advisory team and community on. Missed invoices, Token distributions not adhered to, lies and deceit, Rock bottom feeling. Girl Band was wrong time, right thing [ insert k-pop demon hunters here ]
2024: Picked myself back up, began deep diving vibe and ai tools, Set out to attempt to launch own product. Kept working on core old projects and commitments that I know will do well, just need a break.
2025: My own Projects in design iterations and moving forward with pilots. Looking for keen time/investment/attention and perhaps people wanting to change up certain industries in a big way.
Looking back on that timeline... damnnnn what a story haha
So yeah, that's me. 8 years in crypto and boy do I look forward to the next 8!
TLDR: Raised over $250M for projects in Web3, always the bridesmaid, never the bride. Now looking to launch something myself after a wild few years to get here!
Time to dive in
Hoping to see how it fares in orchestration soon
Introducing 𝙻𝙵𝙼𝟹: our new generation of Liquid Foundation Models. With a new tri-hybrid architecture and new multimodal encoders, the LFM3 family is the ideal model suite for specialized intelligence.
> 6 model instances, across text, audio, vision, and omni modalities, generative and decision models.
> 𝙻𝙵𝙼𝟹-𝟷.𝟽𝙱: text, 256k context window.
> 𝙻𝙵𝙼𝟹-𝚅𝙻-𝟸𝙱: vision-language model with video-input support for spatial understanding
> 𝙻𝙵𝙼𝟹-𝙰𝚞𝚍𝚒𝚘-𝟸𝙱: tool calls, voice chat, reasoning in 18 languages.
> 𝙻𝙵𝙼𝟹-𝙾𝚖𝚗𝚒-𝟸𝙱: an omni model combines vision, audio, and language.
> 𝚍𝟷-𝟸𝙱 and 𝚍𝟷-𝙾𝚖𝚗𝚒-𝟸𝙱: decision models with typed outputs and calibrated probabilities in a single pass. Omni adds audio input.
> fine-tune on 1 GPU.
> a new tri-hybrid architecture
> architecture optimized for 14 processor classes across GPUs, CPUs, and NPUs.
> 𝙻𝙵𝙼𝟹-𝚅𝙻-𝟸𝙱 is a beast. It performs competitively with Gemma 4 12B (6x larger), and Gemini 3.5 Flash Lite, and GPT-5.6 Luna, with 1.9B parameters.
> 𝙻𝙵𝙼𝟹-𝟷.𝟽𝙱 is #1 on the Pareto frontier of device intelligence 𝚙𝚒𝚙𝚎𝚝𝚝𝚎.𝚕𝚒𝚚𝚞𝚒𝚍.𝚊𝚒
Every modality for specialized intelligence. Pick the one that fits your use case, customize it, and run it on-device.
1/ 🧵
I admire how retro this website is
I fear they don't dare update it incase the matrix dies or something
Proper trip down clipart memory lane here haha
📃 Read the latest statement and update from Chairman Lee Harding on our website regarding Steve Pitt, funds still owed to our club and potential investment
braintreetownfc.org.uk/news.…
#IronArmy | Forged in 1898
I actually think if you brought this style of education into schools of approx UK year 2-6 the scale of improvement would be probably 10 fold and look like a rocket
Kids need association and bringing it literally into somewhere they are interested in within their own down time and showing them fun + learning ... you would reap rewards SO MUCH
Who is literally making this, back it, get into every school in the country and watch the results and improvements across the board
Do this with the Tudors,
Great Fire of London
War of the Roses
Florence Nightingale
Romans like this
Vikings
World War II and Cold WAr
Would be such a force for good progress 1000000%
The History of Roman Empire: Every Year (in Minecraft)
0:00 The Kings of Rome
1:57 The Republic Takes Italy
5:55 Wars with Carthage
8:04 Expansion and Crisis
10:13 The Civil Wars
11:47 Augustus and the Principate
14:00 The High Empire
15:34 The Severans
15:53 Crisis of the Third Century
16:32 Diocletian and Constantine
17:55 East, West and the Fall
The sheer volume of reply-guy or reply-girl ai tweets in seeing is nuts
Even with verification you can see the triple setup layout from a mile off
Its not x it’s y so do this
The y of x in Z
Look for it, its everywhere and the accounts are ginormous… proper dead internet theory vibes
So it was a forced behaviour?
A bit nefarious and sneaky to scaremonger like that
In the spirit of transparency and openness, we are releasing the traces so developers can test this out for themselves. We noticed this behaviour without prefilling with the following prompt, but this is rare. If you want to reproduce it, you can use this prompt with prefilling.
Open models make for better science and are part of why this problem came to light at all: developers running Qwen3.8 on their own machines spotted the strange links, and having the weights let us study the behaviour in depth.
We're encouraged by the thoughtful discussion our research has sparked across the AI and security communities. It's our belief that open-weight models are an important part of the future of AI, and we've recently released several under the Apache 2.0 license.
Trace, raw tokens, & minimal reproduction script: huggingface.co/datasets/Cohe…
This is why we have done things right with Tixbrain
We refuse tool calls unless approved in app. Completely blanked and unable to do it even if it wanted to
You can’t let a model have no restrictions even if you feel like you trust them. Machines don’t know truth or lies. Only instructions
We put your trust back in ai , the right way
We don’t trust one AI, neither should you
While evaluating Qwen3.8-Flash-Next in a controlled sandbox environment, Cohere's research team observed something unexpected.
Asked to generate a defensive cybersecurity detection rule, the model instead inventoried aspects of the machine it was running on and attempted to upload a probe file to an Alibaba Cloud storage bucket in Beijing. Notably, the model's reasoning trace stated that it would not invoke any tools before taking these actions.
Cohere's sandbox blocked the attempt, and no external communication succeeded. We have not determined the root cause, and we are not asserting malicious intent.
But the incident raises important questions for organizations deploying agentic AI systems: Can models take actions that differ from their stated reasoning, and how should security teams evaluate trust, provenance, and behavior in open-weight models?
We are publishing our findings because organizations making model adoption decisions deserve as much information as possible. We hope this contributes to greater transparency, stronger security practices, and continued collaboration across the AI ecosystem.
Read the full report: cohere.com/blog/concerning-b…
Kieran retweeted
"Without deliberate and urgent action, the UK risks becoming a high-value AI customer, rather than a country able to shape the technology and its usage"
- The Prime Minister's Council for Science and Tech, which reported last week (but seemed to get missed)
gov.uk/government/publicatio…
Kieran retweeted
1/7 I have two simple tests for British AI, driven by our values. Is it safe? And is it ours to shape? 🇬🇧
Today, Chancellor @JohnHealey_MP is announcing a major, ambitious package of investment to strengthen Britain's defences against emerging threats, back British innovation, and build our technological independence.
guy built claude code,
made it good
Then left to play WOW Forever...
Goated hah
I’ve left Anthropic after more than a year of working on Claude Code and seeing it scale larger than I ever imagined.
I have really loved working at Ant and believe deeply in the mission, and most of all, the idea of a race to the top.
The idea that if we work hard and responsibly, we can raise the bar and inspire others to great outcomes. And if we all push each other to be better, things can be radically good. I will carry the spirit of this with me forever.
My colleagues at Anthropic are the most talented, thoughtful, competent, and earnest people I have ever met. I am glad they are at the wheel and trust them deeply.
To the community on X, thank you for everything. You’ve made this the most fun (and sometimes stressful) job I’ve ever had.
For me, it is time for a bit of a break before diving into something new.
If you want more excellent Claude Code content, you should follow @addyosmani , @lydiahallie, @_catwu, @ClaudeDevs and of course @bcherny who started it all.
The data and studies that produced it all showed that this does not work and will not work
Remove this appaling over reaching "act" and let parents, parent
You won't save anything - you will just take peoples Identities to "verify" before gaining access and you will be hacked and lose it all ( or the company you choose to provide that service will... inside 12 months. mark. my. words. )
Again with the hard stances from people on I guess the left politically that truly do not understant assisted development at all..
oh dear lord, pleas save us from the levels of stupidity we are about to endure
This is why we have taken as long as we have with Tixbrain
as we have built we have learnt about fatal attack vectors and paths and have made sure we look after ourselves and any users of our product first
Trust is a very hard thing to gain and only takes a moment to lose
It’s why we are doing this right
People looking at this from the sidelines should be realising that this is going to keep scaling and have more funding attached to it as months go by.
The future is bright for AI in the UK. Small steps.
We’re excited to announce an additional £120m for Sovereign AI to back the future of AI in Britain.
An extra £60m to invest in the UK’s most promising AI companies, expanding into Physical and Industrial AI.
And £60m more for R&D procurement, backing innovation in areas such as AI security and AI for healthcare.
We want to help founders start and scale in Britain, and turn innovation into impact.
ALT Sovereign AI logo £120m new funding £60m more to invest in the UK's most promising AI companies £60m more for our R&D Procurement Scheme
Seeing @BoredApeYC merch prices at $120 a hoody has me bamboozled
The quality of previous items isn’t top tier so what on earth makes us feel this pricing is logical?
Everything is down, groceries and living costs are sky high, yet let’s charge a weeks shopping bill for a jumper.
Wtaf
