Arthur | Defi KOLs

Been a while we had a hyped TGE. Something new to talk on CT besides hacks, rugs, war and AI. Bullish.

Bitcoin continued to build up last week, making a new recovery high of $78,344. The rally off the $60,033 low could still be described as a bear flag (not unlike the bear market rally last fall), but my sense is that Bitcoin continues to build a large base here in preparation for the next major up wave.

Total degen app on MegaETH. You bet $1 to $50 on a RANDOM long or short for SOL, BTC, or ETH with ~500x leverage. Meaning you don't know if it's short or long when you open. Not much skill involved there. (You can technically figure it out by checking crypto charts quickly, but still) Your actual skill is when to close. Cut losses or wait for the MEGA hit. Hit takes 1% fee on open, 5% of realized profit. Pro perp traders are not the target audience. But it's fresh air with unique implementation and great UI. Shows the direction MegaETH is taking with their ecosystem. Really promising. This is just game one. Options, perps, prediction markets coming. It's not sponsored post but use my ref link for boosted rewards. You earn 2x points and 2x fees on anyone you refer too. NFA: High chance you lose it all so at least do it with my link. And I love earning points. Link: app.hit.one/r/D2DPUK

Got chills down my spine thinking if wBTC got hit with DVN attack. That would've touched every DeFi protocol as well as multiple CEXs who store wBTC.

Keep your X feed clean and crypto maxxed by following these 99 DeFi influencers. Real people with real insights. They're writing daily, so you can learn and earn from crypto and DeFi. There are hundreds more worth following, let us know your favorites 🧠

Crosschain messaging protocols, how competitors compare after Kelp’s $292M exploit 🧠 ↓ 1. Native rollup messaging (Arbitrum's native bridge messaging, OP Stack's CrossDomainMessenger, Linea/zkSync messaging) - The L1 itself is the verifier - Trust inherits entirely from Ethereum consensus - Zero extra committee - The caveat is scope. Canonical bridges only connect a rollup to its L1, not arbitrary chain pairs - The 7-day challenge window is the key UX tradeoff. Most users move to faster options, shifting security from L1 consensus to entity-based quorums → Higher security, often slower, lower liquidity, and limited token support 2. @chainlink CCIP: Separated committees with multiple independent networks cross-check each other - 3 independent oracle networks per transfer: Committing DON, Executing DON, and the Risk Management Network - RMN is built in Rust by a separate team, runs distinct operators, and can halt the system via “curse” transactions - Uses N-version programming (borrowed from aerospace safety) - v1.5 (Jan 2025) added the self-serve CCT standard, so any token issuer can now onboard with their own pool logic and rate limits - No protocol exploits since 2023 If Kelp had been on CCIP, the RMN likely would have caught the supply invariant, even if the primary DONs were compromised by an RPC-poisoning attack. → Chainlink is arguably the most hardened architecture in production today. 3. Single verifier set + economic security (slashing) @axelar - Cosmos SDK-based PoS chain. 75+ validators, DPoS with $AXL staking. Slashing is applied (2% for double-sign, 0.01% for downtime) - Staking concentration in the top 10 validators is a mild concern - Axelar uses Quadratic Voting (shipped in the Maeve upgrade, August 2022) specifically for cross-chain message validation, making it harder for large stakers to unilaterally approve forged messages. - Connects 55+ chains via threshold cryptography - General Message Passing is the messaging layer. ITS is the bridge built on top. @debridge (DMP) - Elected validator set with delegated staking and slashing - Signatures saved to Arweave for verifiability - The validator set is small (historically ~12) and governance-permissioned, which limits decentralization - Unlike Kelp's incident, where LayerZero Labs' DVN relied on compromised third-party RPC providers, deBridge validators run their own full nodes of every supported chain, eliminating the upstream RPC poisoning vector entirely - Battle-tested since 2022 with zero messaging-layer exploits. Sits higher than you'd think because the slashing is real, full nodes are dedicated, and the validator set is actively monitored → Less decentralized, but stronger operational control and economic penalties 4. @wormhole: Multisig-based messaging. Collusion is the risk - 19 Guardian nodes, 13-of-19 multisig quorum. The set is expected to grow over time - Guardians are reputable firms (Jump, Figment, Chorus One, P2P Validator, etc.) running full nodes of all connected chains - After the 2022 $320M exploit: hardened by the Global Accountant (supply-invariant monitoring) and Governor (rate limits +24hr delay on large flows) If Kelp had been on Wormhole, the Global Accountant would likely have flagged $292M in unbacked rsETH before the drain completed → Strong defense-in-depth, but trust ultimately sits with the multisig 5. Modular security frameworks. Safety is configurable @hyperlane - Modular ISMs (apps choose validators + quorum). - Default ISMs on major chains are thin multisigs (2-of-5 on Base, 3-of-7 on Ethereum) - $HYPER token staking via Symbiotic vaults added an economic security layer in April 2025, but application-level ISMs still dominate. 140+ chains, 5 VMs - As safe as whoever wires it up. Renzo adds extra multisig on top of the default ISM for ezETH, which is the right thing to do @LayerZero_Core - DVN-based modular security. Each OApp picks its DVN set, theoretically from 1/1 (trust one entity completely) to 8/12 (institutional grade) - The protocol allowed 1/1 as the quickstart default, which is how Kelp ended up with a single point of failure - Post-Kelp, LayerZero is refusing to sign for any 1/1 configuration, forcing migration. The protocol is technically flexible, which makes the product a footgun → Flexible, but easy to misconfigure. Security depends on apps. 6. @circle's CCTP: Centralized attestation. You trust the protocol entirely. - Circle is the sole attester for USDC burns. There is no multi-sig, no validator network, no cryptographic light client - Works because (a) Circle is a regulated, liability-bearing stablecoin issuer and (b) the burn-and-mint model leaves no honeypot to exploit - As a messaging mechanism in isolation, it's the most centralized thing on this list - For USDC only. Not usable as general-purpose messaging → Most centralized, but effective within its narrow scope Crosschain messaging is a spectrum. On one end: Centralization with restrictions. On the other: fast, flexible systems with added trust assumptions. Other techniques built on top of messaging protocols like cryptographic proofs (@PolyhedraZK, @SuccinctLabs) can eliminate the consensus attack surface, while intent-based bridges shift user risk to solvers. Both can make interop trust-minimized and safer to use.

Must read explanation why BTC lags gold and equities in a risk-on environment. It's no surprise that clarity comes from a TradFi guy, while CT is lost: What's happening to BTC HAS NEVER happened in crypto but happens often in TradFi world: the winning. An IPO moment. Early BTC believers are realizing gains after Bitcoin won. They mined, held through FUD, and now finally have exit liquidity without crashing the market. This is not panic selling but the natural transition from concentrated to distributed ownership. I call it The Great Rotation. When a few thousand wallets hold most of the supply, one seller can move the market. As ownership spreads to millions of investors through ETFs, institutions, and retail, the same amount of BTC sold barely moves price. Selling gets absorbed across time and venues. That is how volatility fades and the asset becomes more stable. Like IPOs, this phase takes time. Early holders sell patiently while new institutional buyers accumulate. It lasts 6 to 18 months, but in Bitcoin time we might already be past the 6 months. This is why this time is different. BTC is not failing to rally because it is a bear market. It is not dumping because of fear. It is consolidating because Bitcoin already won. The market is simply digesting that success. Price moves sideways, sentiment looks dead, but structurally Bitcoin becomes stronger and more mature. OGs exiting is bullish because it means Bitcoin finally reached the scale where large exits no longer break the market. The longer Bitcoin consolidates, the more distributed and resilient it becomes. What feels boring now is how real store of value assets are born.

The Arbitrum Security Council has taken emergency action to freeze the 30,766 ETH being held in the address on Arbitrum One that is connected to the KelpDAO exploit. The Security Council acted with input from law enforcement as to the exploiter’s identity, and, at all times, weighed its commitment to the security and integrity of the Arbitrum community without impacting any Arbitrum users or applications. After significant technical diligence and deliberation, the Security Council identified and executed a technical approach to move funds to safety without affecting any other chain state or Arbitrum users. As of April 20 11:26pm ET the funds have been successfully transferred to an intermediary frozen wallet. They are no longer accessible to the address that originally held the funds, and can only be moved by further action by Arbitrum governance, which will be coordinated with relevant parties.

Arbitrum freezing funds protects L2 users from being treated as second-class in the Kelp recovery. Kelp prioritizing mainnet over bridged assets sets a risky precedent with little upside. Freezing helps rebalance things by showing hacked L2 assets can be stopped. Bridges should add fast freeze mechanisms for hacks too to trap hackers. Yes, it’s more centralized but if you want maximum censorship resistance, use mainnet.

I was checking this same gold chart. But honestly situation and asset is very different. First, the world is wat too much indebt for stagflation to happen - high inflation is the most likely scenario. Second, bitcoin is tiny and not significant asset comparing to gold. So chart should look very different. More vertical up maybe?

I’m shaking

Strategy has acquired 34,164 BTC for ~$2.54 billion at ~$74,395 per bitcoin and has achieved BTC Yield of 9.5% YTD 2026. As of 4/19/2026, we hodl 815,061 $BTC acquired for ~$61.56 billion at ~$75,527 per bitcoin. $MSTR $STRC strategy.com/press/strategy…

@vikktorrrre Let’s bring @PinkBrains_io into the mix as well. 🧠

tier 1 creator agencies every "based" creator would wanna be a part of are 5/5 or 0/5?

@vikktorrrre The agency ecosystem is growing, and each agency is finding its own strengths. At Pink Brains, we focus on DeFi thought leaders on X and genuine, educational content. In the AI content era, that feels more relevant than ever.

33 creator agencies you need to know about

Ethereum Foundation has $48M deposited into Aave on Ethereum. Zero haircut on Ethereum for Aave is definitely a better outcome for EF, which only recently deposited into 'low-risk DeFi' after years of avoiding it. Wonder if this incident changes calculations at the EF.

Kelp rsETH exploit is terrible due to extensive DeFi integrations. Not sure how big the exposure is yet but: - Aave V3: Markets already frozen - SparkLend: Also froze the rsETH market - Lido Earn via Mellow strategy meta-vault. I think it was a leveraged market - Fluid: Frozen market - Compound - Euler - Upshift: Paused High Growth ETH and Kelp Gain vaults - Pendle PT YT tokens - Some Beefy strategies. Yearn? I suppose LayerZero is probably affected too, as rsETH were bridged from L2s, so I wonder if those rsETH on L2s aren't worthless right now. The situation is still developing, so I don't want to FUD any protocol, but it seems there are not many places to hide in DeFi.

@IvanOnTech These are just perma bulls on X. Not impacting the market. Normies are super bearish on Bitcoin now

BULLS HAD A EUPHORIC PREMATURE BULLJACULATION YESTERDAY A lot of emotion Big energy release ALL CHI ENERGY WASTED But - we are in bear trend still Breathe into your balls, stay calm, do not get complacent COLLECT BULL CHI There will be a time to be mega bull! Not yet ..

Optimism might make L2 history here. They're testing paid priority access for transactions. Today every tx competes in a priority gas auction. Pay more, get in sooner. The problem Optimism wants to solve is that this creates spam and gas wars, and traders and market makers can't show commitment beyond gas price. Although this will probably mostly incentivize sandwich attackooors (MEV bots) to stake. Kinda smart. So you can stake 100k OP ($13k USD) and get top-of-block access. - No lockups - instant unstaking - time multiplier prevents quick OP borrowing to frontrun liquidations, you need to hold 15+ days for full boost Phase 1 is first come first served. Phase 2 your stake size, duration, and priority gas creates a single ordering score. This probably first L2 token utility that goes beyond fee revenue (which L2s don't have anyway). plus, Although I wonder if this doesn't end up as sequencer centralization? Because staked sequencer is better than the others. In any case, decentralization is luxury after Base left OP Stack. It was a big hit so $OP needs a real utility narrative and staking is one. Still on testnet, but I welcome innovation in tokenomics. It has been lacking. It seems hard times finally push that innovation.

Roughly 5 years ago, we witnessed a huge influx of VC capital pouring into blockchain and crypto. Now, we're witnessing another historic capital rotation defining the next 5 years. --------- THE WINNERS AI networking & data, memory & storage are the clearest beneficiaries of the AI data center explosion. Broadcom's AI revenue rose 106% YOY in Q1 fiscal 2026, hitting $8.4 billion, with AI networking accounting for about a third and growing 60% YOY. As AI models grow larger, the network connecting thousands of GPUs and the memory storing training data become critical bottlenecks, and massive spending targets. But AI data centers are power-hungry monsters. Quanta Services, with a record $44B backlog, is capitalizing on the convergence of utility grid modernization and the massive power requirements of AI campuses. The grid simply can't keep up with demand, making companies that build power infrastructure and renewable capacity essential enablers. That's where decentralized AI can matter. Distributed, community-owned networks of resources (GPU, data) pose a more energy-sustainable and resilient way to replace centralized AI infras. Aligning with AI infras, the GPU and custom ASIC demand story continues. Nvidia's Q3 fiscal 2026 revenue reached $57B, up 62% YOY. Equipment makers benefit from the downstream effect: every new chip design requires more lithography, packaging, and testing capacity. But infrastructure booms are inherently cyclical. The fiber boom of 1999-2001 is a reminder that capex cycles can correct at any time. Hyperscalers won't spend at this pace forever. Gold has been on a historic run, breaking above $5,000/oz for the first time. Main drivers are geopolitical tensions, DOJ pressure on the Fed that undermined confidence in U.S. institutions, central bank diversification away from the dollar, and expectations of rate cuts. Miners are finally catching up to the metal after lagging for years. Many low-cost operators generate $2,500-$3,000/ounce in free cash flow. Bitcoin has ridden much the same macro tailwinds as gold - inflation hedging, institutional adoption, and de-dollarization narratives - though with more volatility. --------- THE LOSERS While Bitcoin surged as a safe asset, Smart Contract Platforms downing 29% tells the other side of the crypto story. Despite institutional adoption and the rapid growth of tokenization, 6 crypto sectors have declined for the second consecutive quarter, with total fees dropping 30%+. After years of operating at a loss, value accrual is the biggest problem when the music stops. Transaction costs keep falling as cheap L1 and L2 proliferate, meaning more usage doesn't translate to more token value. Macro risk aversion and the broader rotation away from speculative assets compound the pain. Enterprise SaaS and vertical SaaS are suffering the most from the AI wave. Public B2B technology equities have undergone a roughly 25% valuation compression YTD - the sharpest correction since the 2022 rate hikes. The threat only goes bigger in the next few years. If AI agents can do the work of multiple humans, companies need fewer humans, and fewer humans means fewer software seats. AI can be a structural threat to seat-based Salesforce models as customers become more efficient with available tools. Data & Market Intelligence and Cybersecurity are also caught in the crossfire. Claude, Copilot, Gemini, and others can now automate analyst workflows, data aggregation, and threat monitoring better than any mid-senior consultants. The market is questioning what premium these companies deserve. They're hit by the same "seat compression" dynamic, crushing all software. IT Outsourcing is also the flip side of the AI productivity story. Indian IT firms filed more WARN layoff notices in Q1 2026 alone than in all of 2025. With AI delivering more productivity, onsite roles tied to big transformation deals are collapsing. However, heavy tech companies still need human labor while cutting costs. Microsoft announced a $17.5B investment in India's cloud and AI infrastructure after laying off thousands of Americans in 2025. IT outsourcing is being squeezed from both directions: AI is killing traditional human force, but the same cost pressures are pushing new work to outsourcing countries like India and Vietnam, just at a higher skill level. --------- Some final thoughts Although AI is a threat to SaaS companies, deep enterprise integration and data moats may prove more resilient than current pricing suggests. Infrastructure spending cycles are inherently cyclical as resources are limited. And the crypto sector's value-accrual problem isn't solved by more adoption alone. But for now, the market's expectation is priced in. This chart is from @artemis AI. Shout out to @jonbma and his team for this good job.