𝓔𝓶𝓮𝓻𝓪𝓵𝓭 ✧

growth manager/ content writer/web 3🌎 AI✨️Muslim/trader

[𝓔𝓶𝓮𝓻𝓪𝓵𝓭 ✧ @sidiq_fx]

Join the exciting Simply follow BIFU on X to win USDT! on @taskonxyz with me🙌 and START EARNING🤑:
https://t.co/YeWCbsyjqc

[𝓔𝓶𝓮𝓻𝓪𝓵𝓭 ✧ @sidiq_fx]

@UTechStables This is great

[𝓔𝓶𝓮𝓻𝓪𝓵𝓭 ✧ @sidiq_fx]

Moving into the Hood. 7,777 characters on Robinhood Chain.

[PEPEHOOD🏠🐸 @pepe2Hood]

https://t.co/1FITToojbZ

[𝓔𝓶𝓮𝓻𝓪𝓵𝓭 ✧ @sidiq_fx]

Am early for this

[𝓔𝓶𝓮𝓻𝓪𝓵𝓭 ✧ @sidiq_fx]

DeepSafe is building a strong verification layer for the next generation of Web3 and AI. Trustless verification is becoming increasingly important, and this approach could make decentralized systems much more reliable. 🔐🚀

[DeepSafe @DeepSafe_AI]

On September 5, Core DAO published its post-mortem on the reward-accounting exploit that ran from August 28 to 31. Roughly 255M CORE of rewards were issued ahead of schedule.

The cap held. In Core's words, the exploit "did not create any CORE beyond the protocol's limits; the 2.1B maximum supply was never violated." What it did was accelerate rewards the emission schedule would otherwise have released later. What broke was when.

That makes the damage an unusual shape. No user or staking account was drained — Core states that no user or staker funds were lost and that delegated stake was never at risk. What existed instead was surplus, sitting in balances, and it arrived through the ordinary path: the extra rewards "appeared both in attacker-controlled addresses and, unintentionally, in the reward addresses of honest validators who simply received an inflated payout they did not cause."

That made the on-chain reconciliation an arithmetic problem before it was a balance-editing problem. Before any affected balance could be corrected, Core needed a rule for what that balance should have been at the upgrade block.

It used two. Attacker-controlled reward pools were set to zero. For affected honest validators, it defined a fair-earnings floor — the pre-incident balance plus three rounds of that validator's own normal reward — left balances at or below it untouched, and removed only the excess above. Together the reconciliation took 186,153,491 CORE out of supply by direct state reduction. A further ~69M had been moved before the upgrade and sits outside what an on-chain reconciliation can reach; Core says it is pursuing that with law enforcement.

Core says the reconciliation can be checked independently against chain state at the upgrade block, and that the reward paths were subsequently reviewed by Halborn. That part matters as much as the fix. A correction that affected users and independent observers cannot verify is only a second assertion about the balances.

A supply cap is a promise about how much will ever exist. It says nothing about how much should exist yet. Restoring the second one requires knowing a number the cap never had to track.

Sources:
https://t.co/TMhZdwRHaQ
#DeepSafe #CRVA #Web3Security

[𝓔𝓶𝓮𝓻𝓪𝓵𝓭 ✧ @sidiq_fx]

@DeepSafe_AI Strong distinction detection and enforcement are not the same thing

[𝓔𝓶𝓮𝓻𝓪𝓵𝓭 ✧ @sidiq_fx]

@DeepSafe_AI Strong distinction detection and enforcement are not the same thing

[𝓔𝓶𝓮𝓻𝓪𝓵𝓭 ✧ @sidiq_fx]

@DeepSafe_AI @AntalphaGlobal @ViabtcCapital @CogitentV @ShardingCapital @Mason_eagle @Gate @SatoshiLab_HK @CKBEcoFund Strong distinction detection and enforcement are not the same thing

[𝓔𝓶𝓮𝓻𝓪𝓵𝓭 ✧ @sidiq_fx]

@DeepSafe_AI Strong distinction detection and enforcement are not the same thing

[DeepSafe @DeepSafe_AI]

On September 2, TAC published its post-mortem on the August 22 incident that drained its staking pool. The section worth reading twice is not the one about the vulnerability. It is the one about what was watching.

Two controls were running: an automated check comparing TAC's custody against the mirrored supply on BNB Chain and Ethereum, and alerting on large bridge transfers. Both were doing the jobs they were designed to do.

The first had nothing to flag: the attacker locked genuine TAC and issued a correctly backed mirror token, so the conservation property it monitored was never broken. The books stayed balanced.

The second monitored large bridge transfers. Only 32 seconds separated the drain from the first bridge transfer. In TAC's own words: "No control that ends in a human decision closes an interval of that length." That is not a point about how fast anyone noticed. It is about everything that would still have to happen after detection.

This was a drain, not a mint — no new tokens were created. Mainnet resumed on September 2, and the staking pool has since been restored in full.

So the usefulness of a control turns on two properties. Which invariant it watches — because an attack that preserves that invariant stays invisible to it, however correctly the check itself runs. And whether it can act without waiting for someone — because a control that ends in a person inherits that person's response time.

CRVA is designed for a different control position. Where an applicable check is already integrated into a request's execution path and its result is enforced, Agents drawn for that request can return a verdict before the request proceeds. That is not a claim that CRVA would have prevented TAC's internal state-transition flaw. The published record does not establish that an applicable CRVA check was integrated into this execution path. The distinction here is about where a control can act, not about which controls are better.

Both of TAC's controls worked as designed. Neither was an enforcement gate on the transaction that drained the pool.

Sources:

https://t.co/CpY2b8HGZd

#DeepSafe #CRVA #Web3Security

[DeepSafe @DeepSafe_AI]

🎉The DeepSafe Community Quest starts today — a month to explore, create and contribute alongside the community, with 2,000 USDC in rewards.

Whether you have been following DeepSafe for a while or are just joining us, there are plenty of ways to take part.
Complete the mandatory checklist to unlock eligibility, then earn more points by engaging with official posts, sharing your own perspective, creating original content, joining the conversation in Discord and inviting friends.

🏆 2,000 USDC reward pool
📅 September 9 — October 9
📈 More points, a larger share of the pool

Bring your ideas. Make something original.
Start the Quest: https://t.co/7uVaopAafP

#DeepSafe #CRVA

[𝓔𝓶𝓮𝓻𝓪𝓵𝓭 ✧ @sidiq_fx]

AlWeb3 systems still ask us to trust the people verifying the transaction. DeepSafe is taking a different route: cryptographic proof, random verification and hidden committees powered by MPC, ZKP, TEE and Ring-VRF. The goal isn’t more trust it’s less of it. #DeepSafe @DeepSafe_AI

[𝓔𝓶𝓮𝓻𝓪𝓵𝓭 ✧ @sidiq_fx]

minefield.
DeepSafe AI helps bring smarter security to the Web3 world — protecting users from threats before they become problems. 🛡️
Stay safe. Stay ahead.
#DeepSafe @DeepSafe_AI

[DeepSafe @DeepSafe_AI]

DeepSafe is excited to announce our $3m Seed Round with @AntalphaGlobal @ViabtcCapital @capital_spark @CogitentV @ShardingCapital @Mason_eagle @Gate @SatoshiLab_HK @CKBEcoFund

The core decentralized validation technology of DeepSafe has been accepted by the top international cryptographic academic journal IEEE TIFS, with the document ID being 9903072.

Currently, the DeepSafe network has processed nearly 120 million transaction validations, and the number of active accounts on the network exceeds 2.65 million.

Time to make a better Trust Layer now.

[𝓔𝓶𝓮𝓻𝓪𝓵𝓭 ✧ @sidiq_fx]

@DeepSafe_AI the rollback solved one problem but created a bigger question for everyone else

[DeepSafe @DeepSafe_AI]

On September 5, Core DAO published its post-mortem on the reward-accounting exploit that ran from August 28 to 31. Roughly 255M CORE of rewards were issued ahead of schedule.

The cap held. In Core's words, the exploit "did not create any CORE beyond the protocol's limits; the 2.1B maximum supply was never violated." What it did was accelerate rewards the emission schedule would otherwise have released later. What broke was when.

That makes the damage an unusual shape. No user or staking account was drained — Core states that no user or staker funds were lost and that delegated stake was never at risk. What existed instead was surplus, sitting in balances, and it arrived through the ordinary path: the extra rewards "appeared both in attacker-controlled addresses and, unintentionally, in the reward addresses of honest validators who simply received an inflated payout they did not cause."

That made the on-chain reconciliation an arithmetic problem before it was a balance-editing problem. Before any affected balance could be corrected, Core needed a rule for what that balance should have been at the upgrade block.

It used two. Attacker-controlled reward pools were set to zero. For affected honest validators, it defined a fair-earnings floor — the pre-incident balance plus three rounds of that validator's own normal reward — left balances at or below it untouched, and removed only the excess above. Together the reconciliation took 186,153,491 CORE out of supply by direct state reduction. A further ~69M had been moved before the upgrade and sits outside what an on-chain reconciliation can reach; Core says it is pursuing that with law enforcement.

Core says the reconciliation can be checked independently against chain state at the upgrade block, and that the reward paths were subsequently reviewed by Halborn. That part matters as much as the fix. A correction that affected users and independent observers cannot verify is only a second assertion about the balances.

A supply cap is a promise about how much will ever exist. It says nothing about how much should exist yet. Restoring the second one requires knowing a number the cap never had to track.

Sources:
https://t.co/TMhZdwRHaQ
#DeepSafe #CRVA #Web3Security

[DeepSafe @DeepSafe_AI]

Time for the reveal.

Answer: B — one independent decision-maker.

Three valid signatures show that three authorized keys were used. They do not show that control was distributed across three people.

A signer threshold can count keys. By itself, it cannot prove independent control.

[DeepSafe @DeepSafe_AI]

A draft Ethereum standard for AI agents keeps reputation and validation in two separate registries. Both are records. Neither is a guarantee, and the proposal is candid about that.

"Trustless Agents" was created in August 2025 and is still a Draft. It defines three registries: identity, reputation, validation.

The Reputation Registry takes feedback from client addresses — a score, optional tags, pointers to supporting data. It supports on-chain filtering by reviewer and by tag; the proposal expects more complex aggregation to happen off-chain. Its own Security Considerations say the rest plainly: Sybil attacks are possible, inflating the reputation of fake agents. The registry's contribution is to make signals public and put them in one schema.

The Validation Registry records a request for a named validator contract to check a piece of work, followed by that validator's response. The response ranges from 0 to 100 and can represent binary pass/fail or an intermediate outcome. The validator may respond more than once to the same request, enabling updates such as soft and hard finality. It may use stake-secured re-execution, a zkML verifier, a TEE oracle or a trusted judge. The registry is agnostic about which.

Both registries standardise the shape of a trust signal and leave its reliability to the party and procedure behind it. For a registry, that is a reasonable boundary. It also means neither record becomes reliable by being on-chain.

One line sets up the question the proposal leaves open. A validation request must be submitted by the owner or operator of the agent, and the request names the validator address — but the ERC does not say how that validator should be chosen. Incentives and slashing are likewise left to the surrounding validation protocol. CRVA works on that layer: Ring-VRF draws the verifier set for each request, so the set is not fixed in advance. That reduces one selection risk: a set that persists across requests gives an attacker time to prepare. It does not make the check itself correct.

A registry can record who said what. Whether they were right is a different question. Who gets asked is a third one.

Source:

https://t.co/rpBV5ERoya

#DeepSafe #CRVA #Web3Security

[𝓔𝓶𝓮𝓻𝓪𝓵𝓭 ✧ @sidiq_fx]

@DeepSafe_AI the rollback solved one problem but created a bigger question for everyone else

[𝓔𝓶𝓮𝓻𝓪𝓵𝓭 ✧ @sidiq_fx]

@DeepSafe_AI Definitely staying with deep safe