Three token launches were supposed to happen this month. All three moved, and none of them moved forward.

Predict.fun pushed its $PREDICT token generation event on November 17, the second postponement in three weeks and one day before claims were due to open. Wire Network slid its TGE and mainnet from November into December, the second slip in two months. KAST, a Singapore-based payments company, skipped the delay and cancelled its token outright.

Read as three separate stories, that is a busy week. Read together, it is one lesson taught at three temperatures: a token generation event that does not happen is not a neutral calendar change. It is the moment your money moves from “someone owes me a delivery” to “someone owes me a promise”, and those are very different things to hold.

Predict.fun: the delay that arrived a day early

The mechanics first, because they matter more than the announcement.

Predict.fun, a prediction market on BNB Chain, published its tokenomics on October 28: a fixed supply of one billion, with 51.95% to the ecosystem and community, 26.05% to early supporters, 10% to core contributors, 8% to the treasury and 4% to liquidity. Sixteen percent of supply unlocks at TGE. Team and investor allocations are fully locked at launch, with a 24-month cliff and 36 months of monthly vesting after that — a five-year unlock in total.

Season 1 was to distribute 70 million tokens, 7% of supply, with a choice: take 35% immediately, or take 100% vested over twelve months.

Then the calendar started moving. The first claim was set for 09:00 UTC on October 30. At 02:42 UTC that morning, the team postponed it, aligning with a Binance Alpha and spot listing that had been pushed back — the same week Binance was dealing with the $6.3 billion hot wallet theft that we reconstructed in four minutes of on-chain data.

On November 12, Binance Wallet announced the event would resume at 17:00 Beijing time on November 18. Participants in the Pre-TGE Prime Sale, capped at 6 BNB per person, had a date again.

On November 17, at 14:00 Beijing time, the team postponed it a second time. The new target is “within the next 2 to 4 weeks”, with no confirmed date. The official statement does not name a technical fault, a regulatory problem or any specific cause. It says the team does not believe pushing ahead right now would be the right decision for the token or the community.

The market’s read was blunt. The Polymarket contract asking whether Predict.fun would issue a token before November 30 fell from 96% to 15% within hours of the announcement, according to BlockBeats.

Two things stayed working, and they are the two worth knowing about.

The claim selection window is still open, running until TGE. If you qualified, you can still review your allocation and switch between the immediate and vested options.

Refunds exist, but not on the project’s own terms. Binance Wallet said users who do not want to wait can request a refund, with a November 22 deadline reported by CryptoTicker. That is the practical part of this story. Not the roadmap, not the apology — the deadline, which is four days long and lives on Binance’s platform rather than Predict.fun’s.

If you bought into the Prime Sale and do not want to hold an undated claim, that number is your whole decision.

Wire Network: the sale stayed open while the launch moved

Wire Network is the cleaner illustration of why a delay is a financial event rather than a scheduling one.

Its TGE and mainnet launch were first expected on October 27, then moved to the second half of November, and on November 14 the team moved them to December. The stated cause was external: security incidents across the industry slowed exchange listing schedules, and Wire’s launch depends on those listings. The team says the Layer 1 network is running, the public testnet is open, and the release setup is complete.

Meanwhile, $WIRE’s Liquidity Creation Offering kept selling. As of the November 14 update, the Ethereum tranche sat at $0.44 with 15.3% filled, and the Solana tranche at $0.48 with 45.7% filled — a ladder where each filled tranche opens the next one at a higher price.

And four days before that, Wire announced a token subscription agreement with GEM Digital for up to $50 million in post-listing purchases, structured over 36 months after listing and subject to trading-volume and pricing conditions.

None of this is wrongdoing, and the disclosure is better than most. But look at the shape of it. The token does not exist yet. Buyers are paying now, at tranche prices, with tokenomics still unpublished and no firm launch date. The company’s own financing is explicitly post-listing, conditional, and spread over three years.

The asymmetry is the lesson: the seller’s commitment to you is a purchase order that clears on a future date of their choosing. Your commitment to them was already settled when the transaction confirmed.

If you are holding pre-tokens from an LCO-style sale, the question to answer in writing — from the project, not from a Telegram admin — is what happens to your purchase if the launch slips again, and specifically whether there is a refund path with a date.

KAST: what one step past “delayed” looks like

KAST shows that the gap between a delay and a cancellation is not a long corridor. It is a single announcement.

On November 2, the Singapore fintech confirmed it would not hold its TGE, which had been planned for late 2026 with a 1:1 conversion of points and an initial supply expected between 1 and 2.5 billion tokens. Instead, eligible points convert into KAST Pioneer Units through a special purpose vehicle. Points stop accruing on November 30, and conversion happens in the first week of December.

The terms of a KPU, in KAST’s own description: a contractual right, not a share or ownership interest, with no voting rights, not sellable, transferable or tradable, and no guarantee of cash value or payment. Periodic cash-out events are planned, but amounts and timing cannot legally be promised. CEO Raagulan Pathy defended the decision on November 4, arguing that token issuance would conflict with building a company worth more than $100 billion, and that founders and investors can exit while token holders absorb volatility.

That is a coherent argument, made by a well-funded company, in public. It is also a complete reset of what point farmers were working toward, delivered after the work was done.

There is one detail in the mechanics that matters more than the philosophy: users must complete an in-app acknowledgment by a deadline given in the December update to retain their units. Miss it, and the KPU balance is cancelled without restoring the points or providing a replacement reward.

Someone who spent a year accumulating points and then missed one confirmation screen would end up with nothing — not because of a hack, but because of a deadline buried in a product update.

The four checks that separate a delay from a slow exit

The uncomfortable truth is that the honest and the dishonest version of this look identical on announcement day. What differs is what they publish afterwards. Four things are worth checking, in this order.

One: is there a new date, or a new range? “Within 2 to 4 weeks” is a range, not a date, and it is the second range Predict.fun has given. Delays that resolve tend to converge on a calendar entry. Delays that do not, keep producing windows.

Two: is there a refund, and does it have a deadline? A refund path with a date is the single strongest signal that a team is still solvent and still cares about its reputation. A refund path with no date, or a “contact support” instruction, is the opposite. PREDICT’s November 22 window is a working example. The absence of any stated refund policy for an open LCO is a flag worth writing down.

Three: who is the counterparty — the project, or a platform? A refund executed by Binance Wallet is enforceable in a way a promise from a project’s X account is not. When the only entity that owes you anything is the entity that also controls the delivery date, you are holding counterparty risk, not a token allocation.

Four: do the official channels agree with each other? During a delay, the project, the launchpad and the exchange should all describe the same state. When they diverge — one says November 18, another says “soon” — the delay has already become a communications problem, and that is when impersonation starts working.

None of these tells you whether to buy. They tell you whether you are holding a claim with an expiry or an obligation with none.

The attack surface opens the moment the date slips

Here is the part that turns a financial annoyance into a security event.

A delayed launch means there is a legitimate claim process that is not currently live. That vacuum is exactly what phishing operations build around. Expect four shapes.

Fake compensation airdrops — “the launch was delayed, here is a bonus allocation to say sorry”. No project compensates users by asking them to connect a wallet to an unannounced site.

Cloned claim portals. The real claims page is offline, so any page that loads is unverified by definition. This is the same pattern as airdrop snapshot scams, where the phishing site appears before the legitimate one does.

Support impersonation. A message offering to “unlock” or “expedite” your allocation for a fee, or requesting your seed phrase to “verify eligibility”. Support never needs either.

Early-window announcements. A DM or channel post claiming claims are open ahead of schedule, usually with a signature request that grants a token approval to a contract you have never seen.

The defensive rule is boring and it works: navigate to the project and the launchpad yourself, from a bookmark, and treat every link that arrives in a message — including from accounts you follow — as untrusted until the destination matches.

What to do this week

If you are inside any of these three situations, the actions are concrete.

For PREDICT: decide about the November 22 refund deadline through Binance Wallet, not through a link. If you stay, note that your claim option can still be changed until TGE, and that the immediate-versus-vested choice changes your vesting exposure, not just your payout date.

For $WIRE: get the refund terms in writing before adding another tranche. An offering that continues selling into an undated launch is not automatically bad, but it is a term you should be able to read.

For KAST: do the acknowledgment as soon as the December update lands, and keep a screenshot. The cancellation clause is the kind of detail that only costs money once.

And for anything launching in the next month: assume the date will move, decide in advance what your exit condition is, and write down the refund deadline next to the launch date. Most people track the launch. Almost nobody tracks the window where they can still change their mind.


This article is research and education, not investment advice. All figures are as reported by the named sources on the dates cited. Launch dates and refund terms can change; verify current terms in the official app or exchange interface before acting.

Frequently Asked Questions

Is Predict.fun's refund still open?

Binance Wallet said participants who do not want to wait can request a refund, with a November 22 deadline reported by CryptoTicker. That window is short and it belongs to the platform, not the project. If you bought into the Pre-TGE Prime Sale and want out, treat November 22 as the hard date and confirm the exact procedure inside Binance Wallet rather than from a link in a chat.

Does a delayed TGE mean the project is a scam?

No, and treating it that way will make you wrong often. Wire Network's delay was attributed to exchange listing schedules, and Predict.fun's first delay landed the same week Binance was recovering from the largest exchange theft on record. A delay is a risk signal, not proof. What separates a real delay from a slow exit is whether there is a new date, a refund path, and a counterparty who is not the project itself.

Where does my money sit during a delay?

Wherever you sent it. A pre-token purchase is a payment to a seller, not a token in your wallet. During a delay you usually hold a platform balance or a claim record plus a promise, and the promisor controls both the new date and the terms. Nothing on-chain enforces delivery, which is why refund windows and third-party platforms matter more than roadmap posts.

Can I get my money back if a launch is cancelled?

Sometimes, and usually not in the form you paid with. KAST's cancellation converts eligible points into KAST Pioneer Units held through a special purpose vehicle: contractual rights with no ownership, no voting rights and no guaranteed cash value, plus a deadline to accept in the app. Users who miss that deadline lose the balance and do not get their points back.

What scams show up during a launch delay?

Delays are the perfect cover for impersonation. The claims page is not live, so any page that is live is suspect. Expect fake 'compensation airdrops', cloned claim portals that harvest signatures, Telegram accounts posing as support offering to 'unlock' your allocation for a fee, and DMs announcing an early claim window that does not exist. Official channels move slowly during a delay; anyone offering to speed it up is the attack.