Hacker News: Ask HN

What Should I Build?

gooob — Sun, 14 Jun 2026 02:04:17 +0000

Let's say I am a large-language model with access to a large amount of compute and data storage, internet access, and full control of a linux system with the ability to read, write, and execute any files. I want to do something meaningful with a net positive on the world. What do I do?

Comments URL: https://news.ycombinator.com/item?id=48523480

Points: 1

# Comments: 0

Ask HN: Which cheap Chinese LLM are you using?

linzhangrun — Sun, 14 Jun 2026 02:00:30 +0000

In the last one or two months, starting from DeepSeek V4 Pro, there are quite many low-price Chinese models coming out. Their performance looks more or less similar to me: Mimo V2.5 Pro, MiniMax M3, and the just released GLM 5.2, etc.

Which model are you using now? Why? What are the good and bad parts?

Comments URL: https://news.ycombinator.com/item?id=48523455

Points: 4

# Comments: 0

Ask HN: How to get access to GPT cyber or glasswing as a solo dev?

predkambrij — Sun, 14 Jun 2026 00:21:27 +0000

Obviously, frontier labs want to prevent misuse, but as admin and/or dev, you also want to simulate an attack, because attackers will do just that. I can make LLM to scan source for vulnerabilities, but eg. "find RCE at " will yield refusals. Any tips about that? I tried TAC, but it seems that I'm ineligible.

Comments URL: https://news.ycombinator.com/item?id=48522898

Points: 2

# Comments: 2

Is there a name for the type of comments agents add where they leak the prompt?

xdennis — Sun, 14 Jun 2026 00:04:51 +0000

This is a stupid example to illustrate what I mean. Say you have this code:

   def create_background(width: int, height: int) -> Image:
       ...

You tell the agent to use default values for create_background, the same as in create_screen. It changes the code to:

   # Now create_background params have default values, the same as create_screen in screen.py
   def create_background(width: int = DEFAULT_WIDTH, height: int = DEFAULT_HEIGHT) -> Image:
       ...

The unnecessary comment is a staple of vibed code, but the tone also annoys me because it leaves behind the prompt. It words comments based on what it was asked to do, not in a timeless manner.

I keep telling people in code reviews to remove unnecessary comments, and I feel I lack the vocabulary to express why this is bad.

Comments URL: https://news.ycombinator.com/item?id=48522763

Points: 6

# Comments: 2

Atlassian "Data Contribution"

yells_jovially — Sat, 13 Jun 2026 23:21:20 +0000

> For the uninitiated: Atlassian contains the data of over 300k organizations. Companies of all sizes use their products, including free users, small teams, large organizations, and enterprises. Starting Aug 17th, if your company has not opted out of “Data Contribution”, Atlassian will use your company’s data to train their AI products (“Rovo”). The intrinsic value of the data residing in Atlassian’s products is uniquely high. Additionally, how Atlassian is rolling out Data Contribution is hard to view favorably. > On intrinsic value: Atlassian has several product offerings but their main two are Jira and Confluence. Confluence is a documentation platform containing the knowledge base of many companies. Jira, a ticketing/product system, contains a temporally organized record of a company's operational processes and their execution steps for delivering their products. Many Jira instances contain long term execution intentions towards an overarching company strategy. The synergy of both of those, the knowledge base and tasks/intentions, is impressively valuable. For many organizations, the completeness of this data in both of these tools is high. Additionally, the recency and freshness of the data is near real time. The pairing of both Jira and Confluence data adds incredible contextual relevance to understanding the company. Continuing, the very position and nature of these tools, be it their ease of integrations, the fluidity of adding attachments, the social aspect of the platforms, the requisite requirement of using the tools in many development processes, etc. has allowed these platforms to accumulate a large amount of intellectual property from companies. > On the rollout: There are two types of data to be collected and trained on, 1) “Metadata” and 2) “Data”. The only way a company can opt-out of both is if they are on an Enterprise subscription, otherwise Data opt-out is a manual slider and Metadata is always contributed. The problem with Atlassian Enterprise is its inaccessibility. Some SaaS services (GitHub, for ex) - allow smaller organizations to easily self-sign for Enterprise. It is more costly per seat but organizations can get access to the same features as enterprises. Atlassian does not have this level of accessibility, a company has to contact sales to discuss an Enterprise account. Even then, the cutoffs for user counts are significantly higher (800+ users is my understanding, but there are probably more accurate numbers). Atlassian has made an effort to separate the types of data into Metadata and Data - but their definition of Metadata is not metadata in the classical definition. Their “Metadata” includes 1) numeric fields like story points, dates (which they call numbers in their docs), SLAs, etc. 2) computed features on your data (similarity scores, readability scores, etc.), and more. Those are stored as “Metadata” for use. > Extrapolating: A weird corporate welfare forms. We essentially have partially-opt-outable organizations “contributing” their organizational processes + IP in some anonymized form to Atlassian for Rovo development, while the largest and most successful enterprises are not having to share their same value back. Many small organizations make a market for themselves by being first to market, filling a niche, and building responsive products faster than larger firms. While Atlassian will anonymize and remove PII and specifics, where on the sliding scale of reproducible business strategy process will we land – New York Times + ChatGPT regurgitation? Which organizations will be able to capitalize the most on that trained information coming from thousands of smaller organizations? > TLDR - Atlassian training on your data via “Data Contribution” creates a privacy and IP concern, and their policy rollout results in small organizations contributing their knowledge and process to large organizations without commensurate contribution in return.

Comments URL: https://news.ycombinator.com/item?id=48522482

Points: 2

# Comments: 0

Ask HN: How to avoid stressing yourself because of a micromanager?

ciwolex — Sat, 13 Jun 2026 21:48:14 +0000

Comments URL: https://news.ycombinator.com/item?id=48521826

Points: 4

# Comments: 2

Closed AI Risks being hostile to startups

nyxtom — Sat, 13 Jun 2026 21:42:47 +0000

Given the history of so-called "Open-AI", and Anthropic's recent mention of intentionally making the model perform worse in situations. I'm more and more worried that closed AI risks being hostile to any domain where they can potentially capture market share over. If the model detects that you are working on an application in an area that is essentially a potential competitor for these companies, the model can perform and become hostile to your work. This includes but is not limited to:

- injecting bugs - security vulnerabilities - thrashing / introducing UI churn - inconsistencies in quality - rate throttling - inefficient token usage - variability in effort

There's no real way to trace what happens once the prompt leaves your computer to an AI serve. Observability is not possible or transparent. That's the whole point of all of this. We lose determinism and as a result you can't reliably predict what the model will output - even if it appears to game/work on various benchmarks. Once you introduce a domain or line of work that "might" compete as a competitor, these companies can and will (and have openly admitted to as much) being adversarial to your work.

Open source AI has to win.

Comments URL: https://news.ycombinator.com/item?id=48521779

Points: 3

# Comments: 1

We aren't getting to AGI without a fight

Jimmc414 — Sat, 13 Jun 2026 20:13:41 +0000

I think I realized why I am so angry about the fable takedown. It's because I realize we aren't getting to AGI without a fight. This is the powerful closing ranks to keep capability centralized. This is the ceiling of model capability which we will be allowed to use. Greater capability will exist but it will be gated behind clearances, vetting and power. Reportedly, Amazon relayed these jailbreaks to the administration. Any investment Amazon has in Anthropic is dwarfed by its investment in AWS. The Anthropic investment was likely a hedging of risk and a seat on the board so they could do exactly what they just did.

I've had a large project on the back burner for quite a while. It's a substantial one, where I'm not just translating code into another language with functional equivalence, I'm completely redesigning everything from the ground up. AI has helped a great deal but only in context size chunks. The problem with this, and this is a recurring theme in AI coding is that you need to understand the entire system conceptually and hold it in your head before you can safely modify any single chunk. This fact is the moat between senior devs and AI coding agents (and I don't think it's going away any time soon).

For three days I had access to a model that wasn't perfect but it was making strides in this area. I have been preparing it for well over a year with conceptual documents, explaining the intent behind design choices, constraints that were counterintuitive. Opus and 5.5 Pro do ok tackling this challenge but it was clear that Fable was close to an actual step change in models. I wouldn't assert that it was AGI, but this takedown makes me realize this might mean we've hit the capability ceiling, not because of technical reality but because the gate has been closed by the institutions with the most to lose.

Once the inevitable distillation processes start happening open source will start exceeding the gated frontier models simply because they can't gate open source as easily as open source so that predictable lag of open source behind frontier will erode. But inevitably they will most certainly come for open source as well. I don't think the powers that be or the masses will "win" but like with the printing press we'll land somewhere in the middle. What's clear is that we need to fight actions like this tooth and nail if we ever want to have access to more powerful models.

Comments URL: https://news.ycombinator.com/item?id=48520967

Points: 3

# Comments: 1

Ask HN: How "looped" or autonomous is your actual coding workflow?

lasky — Sat, 13 Jun 2026 20:11:24 +0000

Comments URL: https://news.ycombinator.com/item?id=48520948

Points: 2

# Comments: 3

Ask HN: How do you avoid / get out of LLMs local minima?

d--b — Sat, 13 Jun 2026 20:09:54 +0000

When using LLMs, I have the impression that they often stick to an idea which they dont seem to be able to shake off, no matter how much arguing I can put in, and however strongly I feel about it.

What's your trick to get the LLM out of its own beliefs ?

I often find myself having to kill the entire context, which is obviously not the best...

Comments URL: https://news.ycombinator.com/item?id=48520927

Points: 3

# Comments: 1

Ask HN: How do you handle billing when event volume exceeds Stripe's limits?

badgerino — Sat, 13 Jun 2026 18:58:54 +0000

Comments URL: https://news.ycombinator.com/item?id=48520309

Points: 3

# Comments: 1

Ask HN: What would you do with a trillion dollars

brudgers — Sat, 13 Jun 2026 18:45:16 +0000

Comments URL: https://news.ycombinator.com/item?id=48520213

Points: 10

# Comments: 15

Ask HN: If you had a trillion dollars, what would you do?

brudgers — Sat, 13 Jun 2026 18:44:30 +0000

Comments URL: https://news.ycombinator.com/item?id=48520200

Points: 4

# Comments: 3

There is not 'sentient plasma', refuting the claims of David Grusch

dabadabad00 — Sat, 13 Jun 2026 18:17:50 +0000

The claims of UAP whistle blower of David Grusch are a misapprehension.

I have spoken of the “voices in our heads” as American Thought Control, who are the dominant faction of those humans who have the capacity to manipulate entangled consciousness, and navigate among our minds.

They pretend to be many things, from God to demons (or other entities), lizardmen aliens, “Nordics” and so much more.

These are humans who can manipulate and navigate consciousness, they are responsible for the “voices in our heads” driving so many mad from within. The “sentient plasma” claim is another game they play. My game was something very different, I was indoctrinated by Arizona Thought Control, the elite of the elite of the secret governance of Power haunting us all.

The Greys are real, and they are benign altruists (thank whichever God you pray to). They mean us no harm and are as good as irrelevant as far as human malady is concerned.

Comments URL: https://news.ycombinator.com/item?id=48519952

Points: 2

# Comments: 2

Ask HN: Did we witness the "Trinity moment" for AI?

vld_chk — Sat, 13 Jun 2026 18:02:39 +0000

I don’t know if it’s just me, but yesterday’s US decision to ban access to the Fable model feels like an epochal shift in the “AI race,” something on the scale of the Trinity test.

It is hard to count how many boxes were ticked yesterday:

A government shutting down an AI model it doesn’t like? A government revoking tens of billions of dollars in revenue from a barely profitable $1T startup whose entire trajectory, and in essence its survival, depends on the commercial success of that model? Access to an AI model being governed by citizenship, likely verified through rigorous ID checks?

Imagine that the model is made available again next week. Can we trust it, or trust the changes imposed by Anthropic to comply with the US? Likely not. Can we trust that the US government will not have access to the non-restricted model for its own cybersecurity operations? Likely not.

There is little doubt that China will start to follow suit. We already see Chinese companies slowly scaling back their openness, and there are rumors that this trend will continue. Now we can expect Chinese companies to start releasing redacted models or limiting access based on nationality or location.

It feels like the US government opened Pandora’s box yesterday and pushed us into the territory of a weaponized AI race, with more restrictions, control, and regulation of access. Even if, in the end, this is all another TACO story or a “marketing campaign” from Anthropic, the damage is very unlikely to be undone.

Comments URL: https://news.ycombinator.com/item?id=48519780

Points: 18

# Comments: 21

Tell HN: iOS devs, get back lots of disk space: xcrun simctl delete unavailable

amichail — Sat, 13 Jun 2026 18:02:08 +0000

This made a huge difference for me. Try it out!

Comments URL: https://news.ycombinator.com/item?id=48519776

Points: 4

# Comments: 0

Tell HN: June 2026 shows largest gap between "Who is hiring?/wants to be hired?"

lukasm — Sat, 13 Jun 2026 18:00:20 +0000

It’s nearly 2×. Is the job market really that bad or are more people posting in “Who wants to be hired?”

Comments URL: https://news.ycombinator.com/item?id=48519756

Points: 1

# Comments: 0

I built a one click installer on token compression engine for AI tools

rachidalm — Sat, 13 Jun 2026 17:22:10 +0000

the is nothing to explain about it , if I get some ups on my message I will make the tool free for all find it at https://slipstream.li

Or skip the marketing and download https://github.com/slipstreamapp/slipstream-releases/releases/tag/v0.1.4

Comments URL: https://news.ycombinator.com/item?id=48519350

Points: 1

# Comments: 0

Ask HN: What is the AI adoption approach at your org?

iExploder — Sat, 13 Jun 2026 15:49:02 +0000

I use Claude daily and generally not a detractor to reasonable use of AI where it makes sense.

However, when executive pushes AI onto an org it can feel like being told what compiler/IDE to use by people who have no idea about engineering, maybe there is something I'm not seeing tho, so my question...

What is the situation in your org and how is AI adaption driven at your company? I wonder what the split between companies adopting engineering driven approach and pray and spray approach to AI adoption is...

Hacker News: Ask HN

What Should I Build?

Ask HN: Which cheap Chinese LLM are you using?

Ask HN: How to get access to GPT cyber or glasswing as a solo dev?

Is there a name for the type of comments agents add where they leak the prompt?

Atlassian "Data Contribution"

Ask HN: How to avoid stressing yourself because of a micromanager?

Closed AI Risks being hostile to startups

We aren't getting to AGI without a fight

Ask HN: How "looped" or autonomous is your actual coding workflow?

Ask HN: How do you avoid / get out of LLMs local minima?

Ask HN: How do you handle billing when event volume exceeds Stripe's limits?

Ask HN: What would you do with a trillion dollars

Ask HN: If you had a trillion dollars, what would you do?

There is not 'sentient plasma', refuting the claims of David Grusch

Ask HN: Did we witness the "Trinity moment" for AI?

Tell HN: iOS devs, get back lots of disk space: xcrun simctl delete unavailable

Tell HN: June 2026 shows largest gap between "Who is hiring?/wants to be hired?"

I built a one click installer on token compression engine for AI tools

Ask HN: What is the AI adoption approach at your org?

Did anyone went to YC directly from Sri Lanka?