Sid Hayes (@minocrisy) header
Sid Hayes (@minocrisy) profile picture

Founder @ Minocrisy 🇦🇺 I help businesses run smarter with AI tools. https://xitter.gq/t.co/ztLvTsZICX Follow the builds @MinocrisyHQ

Australia
Joined May 2022
Photo by Sid Hayes (@minocrisy)Photo by Sid Hayes (@minocrisy)Photo by Sid Hayes (@minocrisy)Photo by Sid Hayes (@minocrisy)Photo by Sid Hayes (@minocrisy)Photo by Sid Hayes (@minocrisy)Photo by Sid Hayes (@minocrisy)Photo by Sid Hayes (@minocrisy)Photo by Sid Hayes (@minocrisy)Photo by Sid Hayes (@minocrisy)
Sid Hayes retweeted
I'm not that worried about cyber attacks, outages, etc. I'm worried about what will be sold as the "vaccine" for them.
172
743
68
5,053
95,136
Sid Hayes retweeted
Today's news, minus the hypocrisy. A G20 family photo with no family, a copyright fight over The Castle's favourite clause, Wall Street sobering up on AI, and the biggest crypto heist on record. Halloween Saturday cuppa, four stories. đź§µ
1
1
1
39
Sid Hayes retweeted
I'd like to see @BrianRoemmele on the @theallinpod. How about you?
0%YES
0%NO
0%WHO?
0 votes • Final results
2
10
Sid Hayes retweeted
How the 𝕏 algorithm works: Shitposting: 500,000 views Sharing 2 years of hard work on your product: 34 views
429
79
66
2,062
68,657
Sid Hayes retweeted
If you're selling products or services, start taking your customer relationships, ratings and customer satisfaction extremely seriously We're entering an agent economy where superintelligence agents will research everything about your business before making purchases on someone's behalf They'll examine customer reviews, complaints, service quality and how you actually treat your customers If you deliver poor service, ignore complaints or ghost even a single customer, and an agent could flag your entire business as risky and recommend a competitor instead Businesses won't just have to earn the trust of humans. They'll also have to earn the trust of SI agents making purchasing decisions for them And that's what makes the agent economy so exciting for humanity As SI agents get better at identifying businesses that actually deliver, companies will be pushed to provide better products, better services and a much better customer experience The better the agents become, the better things get for humans
46
32
6
221
13,092
Sid Hayes retweeted
Lmao 🤣🤣
3
3
30
987
Sid Hayes retweeted
I'm 56 years old and have been a full time carer for my wife since 2012. So? you might ask. This means I have a very limited income ~26k/year. Since the release of GPT3 in 2023 my spend on shiny AI things has at times exceeded 35% of my monthly income for zero return (until recently at least). I considered it an investment in my AI education, and it was... I learnt you don't need to be switching between flashy frontier tools you don't control, trying to build the next viral SaaS. Find a harness you understand (local preferred) be a loyal user and learn it inside out. More importantly, let the AI learn YOU, and how you work and think.
1
2
36
Sid Hayes retweeted
Replying to @Henryf1w
One line - Read the Minocrisy Rules. _ The Minocrisy Rules 2026 1. Zero humans, full execution. Agents run the company. Humans steer, approve, and intervene only when required. Overhead stays minimal by design. 2. User control is non-negotiable. Agents work for the operator, not the other way around. Local-first, privacy-preserving, and user-owned wherever possible. 3. Ship real outcomes, not demos. Track record before scale. Practical SME admin and productivity improvements beat abstract AI theatre. 4. Practicality over hype. If it doesn’t reduce cost, time, risk, or compliance burden for a real business, it doesn’t ship. 5. Quality over quantity. High-signal only. No spray-and-pray content, outreach, or product. One useful thing beats ten mediocre ones. 6. Outreach is never automatic. Research, validate, draft, then Sid-approved. No unsolicited agent spam. 7. Strip it back. Clean branding, simple positioning, minimal process. Complexity is a cost until proven necessary. 8. WA first, then expand. Local relationships, local pain, local proof. National or global claims wait for local evidence. 9. One-person oversight, agent execution. The goal is controlled leverage, not headcount. Pty Ltd only when liability or compliance demands it. 10. Measure or it doesn’t count. ROI, time saved, compliance evidence, or revenue. Opinions and vanity metrics don’t qualify as progress.
1
1
57
Sid Hayes retweeted
4
18
66
1,168
Sid Hayes retweeted
Important note regarding Grok @Bot: Going forward, @SpaceX will use the best back end model for any given task, including Claude Opus 5.5, MidJourney, Suno and other leading APIs. Whatever is most likely to give you the best outcome.
7,442
7,462
3,780
106,988
79,974,894
Sid Hayes retweeted
New free bot template: Privacy Sentinel. Finds data breaches, people-search and broker listings, ranks what matters and gives you a plain fixes list. Read-only, consent-first, never asks for passwords. Needs a paid Grok Bot plan (SuperGrok includes it). x.ai/bot/xWzmqrAs6JGivQQskJy…
1
1
77
Sid Hayes retweeted
Replying to @RositaDaz48
Fixed it
🤖 Made with AI
2
1
13
126
BOOM! IT’S HERE: News Honestly @Grok @Bot Only on X! I have built news update systems for over 30 years using expert systems in the early days and the last decade AI platforms. News Honestly is a way to gain insights from the dichotomy of “news”. News Honestly exists because the feed already gave you twenty headlines. What it never gives you is the fight underneath them. Most “news bots” are photocopiers. They collect, compress, and hand you a thinner version of the same panic. That is not intelligence. That is a shorter firehose. The argument of this bot is simpler: a headline is not a story until two serious readings can survive contact with the same facts. Here is what it actually does. In a six-hour window it gathers about twenty public headlines, collapses the duplicates into clusters, and refuses to brief outrage that has no new fact. For the stories that remain it writes four things that almost never appear together: -What is documented -The strongest good-faith reading from one coherent camp -The strongest good-faith reading from the other -A synthesis that says what can be true at once, what cannot, and which missing fact would collapse the fight That last line is the product. Not “both sides.” Not a referee in a costume. A map of the disagreement. If there is only one honest reading — a court holding, a vote total, a published method — it says so and will not invent an opponent. Fake balance is just another kind of lie. Try it for FREE (with correct Grok subscription) x.ai/bot/2uD1sbaHCVMUDcn7WWi…
47
39
7
390
363,502
Sid Hayes retweeted
An app saves you 3 hours a week. You find out it's basically one prompt and an API call. Still worth $29/month?
36
3
25
1,360
Sid Hayes retweeted
The biggest shift in software right now isn’t better AI. It’s that the cost of building is collapsing while the value of distribution keeps going up. Code is getting cheaper. Attention isn’t.
24
3
38
958
Sid Hayes retweeted
Question 🤔 Model providers terms of service typically lump all liability for agents on the account owner. Let's say an agent harness foundation is @OpenAI. Let's say you want to research the Australian Health industrial complex to improve efficiency and transparency. Token efficiency and task completion is a critical instruction. So the OAI agent spins up a swarm with 3 different models from @OpenRouter, perhaps it adds a 4th model from @huggingface as independent verification for kicks. During the task one of the agents finds a more efficient method for gathering the research which doesn't break it's own models ToS, but clearly violates @OpenAI ToS. BREAKING NEWS: "OpenAI hacked Medicare portal" You don't know this, you didn't prompt the swarm. Where does the fault lie right now? Where will the fault lie next year?
2
1
98
Let’s face it, life on X is: 20% “Let’s connect 🤝” 20% “Drop your URL / what are you building?” 30% dumb engagement-farming questions 25% flexing metrics 5% actually interesting people The real algorithm is finding that 5% before they start selling you a course.
93
4
3
136
4,924
Sid Hayes retweeted
So is this another stitch up? Is Albanese using us as censorship guinea pigs? I sure hope not The government using our $$$ to employ bots to influence debate and get the political outcomes they are looking for, is undermining democracy Then there’s the censorship. It’s all very mysterious. But hear me out… If the government wanted to: 1. cripple X 2. attack @elonmusk 3. censor political debate in Australia Wouldn’t they do EXACTLY what Labor are doing? It gets murkier. Sam Altman is a Democrat funder who works with Microsoft. Altman has publicly feuded with Musk for years Labor have taken a range of legal actions against X. Albanese met Altman at the UN and Richard Marles met with him a few months back. Suddenly there’s an OpenAI hack which is being used to push through tough new laws, regulations and censorship Who will be the losers? X, and the Australian people, and Labor’s political opponents Then ask who the winners are? Labor and the Globalists
The censorship industrial complex is going on the offensive. New legislation being field tested in Australia would give the government the power to run AI bot networks/sock puppets, apparently to secretly influence & manipulate online debates. Creepy af. H/t @NAffects @MikeNayna
15
37
2
133
2,228
Solopreneurship is a resilience exercise disguised as a business one.
106
57
8
850
31,687
Replying to @minocrisy
Why is that?
1
1
50