Twitter’s approach of blacklisting specific InfoFi apps is argued to be ineffective because similar spam in
Key takeaways
- Twitter’s approach of blacklisting specific InfoFi apps is argued to be ineffective because similar spam incentives can reappear in new forms and require repeated blacklisting.
- The corpus presents a tension that platforms position prediction contracts as distinct from gambling for regulatory advantage even though much user activity is effectively gambling.
- Lighter has bought back about 1.4 million LIT (about $2.8 million) since December 29.
- A stated dispute is that a crypto-native marketing agency could be a strong business for a small team but may not justify a publicly traded token valuation.
- A stated dispute is that social/copy-trading features are widely believed to be a slam-dunk but historically have not become the dominant venue because trade-sharing already happens on existing social platforms.
Sections
Social Platforms: Attacker Economics, Engagement Incentives, And User-Side Filtering
The deltas emphasize that spam/bot problems are rooted in incentive gradients: low-cost content creation with high potential payout, plus engagement-ranked feeds. Verification and subscriptions are portrayed as insufficient signals of authenticity and possibly misaligned with enforcement incentives. The operational burden is described as shifting toward users via blocking/muting, with tools like extensions described as inadequate. A separate expectation is that creator migration to long-form channels may be underway, but the corpus also argues spam dynamics recur across platforms over time.
- Twitter’s approach of blacklisting specific InfoFi apps is argued to be ineffective because similar spam incentives can reappear in new forms and require repeated blacklisting.
- Bot and scam content on Twitter may be fundamentally unsolvable given the platform design and incentives.
- To avoid seeing low-quality content on Twitter, users must persistently block and mute accounts and keywords.
- Recommendation algorithms on Twitter are described as highly sensitive to clicks and dwell time, so engaging with undesirable posts causes the system to show more similar posts.
- Spam is described as a recurring pattern across communication eras (phone calls, email, social media) requiring ongoing user-side filtering like blocking and unsubscribing.
- SPEAKER_01 reports not using Twitter for two months aside from tweeting once or twice.
Prediction Markets: Yield-Like Marketing, Litigation, And Regulatory Classification Risk
The corpus signals a shift in how prediction markets are being positioned: some participants market near-certain contracts as bond-like yield instruments, while a dispute argues the risk framing is misleading due to tail events. A concrete legal development (Kalshi lawsuit over an ad portraying predictions as financial help) indicates consumer-protection/regulatory sensitivity around marketing. Regulatory arbitrage is described as central—emphasizing contract form to avoid “gambling” classification—while the corpus acknowledges dual-use behavior (gambling vs financial use) and notes sports markets are clearly gambling and financially important, constraining product scope changes. A multi-year timeline for regulatory clarity is presented as an expectation.
- The corpus presents a tension that platforms position prediction contracts as distinct from gambling for regulatory advantage even though much user activity is effectively gambling.
- There is increasing need for clearer rules requiring influencers to disclose whether and how much they are paid when promoting financial strategies or platforms, even when platforms are not directly paying them.
- Kalshi was sued over a TikTok ad that portrayed the product as a financial tool by implying users could pay rent after winning on predictions.
- A stated dispute is that marketing prediction markets as risk-free “bond-like” instruments is misleading because stacking many binary bets can still result in total loss when an unexpected outcome occurs.
- Prediction market operators attempt to position themselves as “predictions” rather than “gambling” to obtain a regulatory advantage even though the underlying activity is still gambling.
- Some prediction-market participants promote “prediction market bonds” aimed at capturing small mispricings on near-certain outcomes and claim theoretically up to about 40% APR.
Perps Dex Competition, Distribution, And Token Value Capture
The corpus highlights a post-launch reality check for perps venues: measurable buybacks exist alongside evidence of declining volumes/market share for Lighter, and a claim that sector-wide volumes are down. Product distribution (Lighter mobile app vs Hyperliquid lacking native mobile) is presented as a competitive lever. Monetization and value capture are framed as changing via a zero-fee, API/PFOF-like approach, which alters how to interpret volume as revenue. Airdrop holder turnover is flagged as potentially largely complete but measurement is uncertain.
- Lighter has bought back about 1.4 million LIT (about $2.8 million) since December 29.
- Hyperliquid does not have a native mobile app, and its iPhone Safari progressive web app experience is described as usable.
- Lighter’s perps activity and market share declined from a peak around November 21 to roughly about $1.2 million per week recently.
- Perps volumes have been declining across the sector since mid-2024, including Hyperliquid.
- Only about 37% of Lighter’s airdrop is still held according to a tracker estimate.
- The tracker estimate for Lighter airdrop retention may miscount transfers as sales.
Kaito/Kaido: Pivot From Permissionless Engagement Mining To Curated Marketing
The corpus presents a clean before/after: incentives tied to engagement and allocations are described as driving botting, followed by an explicit announcement that Yaps is being sunset in favor of a curated tiered marketing model. This reframes the product from open participation toward a vetted influencer/campaign workflow with analytics. Token/business-model fit is challenged via a dispute framing the resulting business as potentially agency-like, and a specific on-chain timing watch item (unstaking shortly before the announcement) is raised as a trust risk but not proven.
- A stated dispute is that a crypto-native marketing agency could be a strong business for a small team but may not justify a publicly traded token valuation.
- A large Kaido unstaking occurred about four days before the program-sunsetting announcement, raising suspicion of advance knowledge (while noting it could be coincidence).
- Kaito announced it is sunsetting the Yaps program and shifting toward a curated, tiered marketing model connecting vetted key opinion leaders with campaigns and analytics tooling.
- Kaido’s fully diluted valuation dropped to roughly $550–$560 million following commentary around a Nikita tweet.
- Kaito farming and botting were driven by incentives where staking Kaito tokens and/or being an active “yapper” could influence allocations into sales/raises hosted on Kaito’s platform.
- Ryan’s original thesis was that Kaido aimed to disintermediate marketing agencies by enabling teams to run scalable guerilla influencer campaigns through a self-serve platform.
Pump: In-App Social Trading Features Versus Bot-Driven Microstructure
Pump is described as adding an in-app “call outs” social layer while simultaneously facing a known microstructure failure mode: bots snipe explicit calls and offload to slower participants. Revenue is reported as recently rising, which is a traction indicator but not linked to the new feature within the corpus. The corpus also disputes that social/copy trading is automatically a winner, arguing trade-sharing already happens elsewhere. Mitigations like gating/KYC are presented as imperfect due to identity rental/purchase.
- A stated dispute is that social/copy-trading features are widely believed to be a slam-dunk but historically have not become the dominant venue because trade-sharing already happens on existing social platforms.
- Explicit on-chain token calls tend to be gamed by bots that automatically snipe the ticker and then drip supply to manual buyers.
- Pump’s revenue rose from around $1.0 million in late December to roughly $1.6–$1.8 million over the past two days.
- Pump’s new “call outs” feature allows traders to broadcast token calls directly to followers inside the app.
- Platforms pursue social trade-sharing because it can create vendor lock-in and increase a trading flywheel that drives revenue if users copy trades quickly.
- Preventing bot exploitation in live microcap token calls likely requires gating who can trade, but gating can be circumvented through purchased or rented KYC identities.
Watchlist
- A large Kaido unstaking occurred about four days before the program-sunsetting announcement, raising suspicion of advance knowledge (while noting it could be coincidence).
- There is increasing need for clearer rules requiring influencers to disclose whether and how much they are paid when promoting financial strategies or platforms, even when platforms are not directly paying them.
- Active policy discussions include whether stablecoins will be allowed to pay yield and how tokenized equities will be regulated.
Unknowns
- What portion of Lighter’s buybacks is funded by sustainable protocol revenue versus treasury/other sources?
- How much of Lighter’s volume decline is explained by overall sector contraction versus competitive share loss, and over what exact time windows?
- What are the actual revenue mechanics and magnitude for Lighter’s API/PFOF-like monetization, and how do they scale with volume?
- Did Lighter’s mobile app launch measurably change acquisition, retention, or retail share of volume?
- What is the true post-airdrop sell/hold behavior for Lighter once transfers are separated from sales and exchange inflows are measured?