The 4 Truths and Fee Traps Behind Polymarket's LP Market Making Incentives

marsbitDipublikasikan tanggal 2026-03-22Terakhir diperbarui pada 2026-03-22

Abstrak

Polymarket, a prediction market platform, has recently shifted its focus to incentivizing liquidity providers (LPs) to address its core issue of low liquidity. While most markets remain free, it now charges a taker fee on specific markets like crypto price movements and select sports events. This fee, highest near 50% probability, funds new LP reward programs. There are two primary reward systems: one pays LPs when their limit orders are executed (maker rewards), and another rewards simply for placing orders within a set spread to provide liquidity, even if they don't get filled. A third mechanism allows anyone to sponsor additional incentives for specific markets. A positive view argues this structure values genuine liquidity over mere trading volume, making fees earned and rewards received a potential key, anti-sybil metric for a future POLY token airdrop. It rewards users who improve market depth and stability. A contrasting, negative view claims the LP program is a "trap." Critics argue that professional market makers avoid it due to insider trading risks and that most LPs are actually losing money due to hidden "LP wear and tear" (impermanent loss), only participating based on speculation of a valuable airdrop. They warn that if Polymarket expands fees to fund these unsustainable rewards, it could lose its competitive edge of zero fees and better odds compared to traditional sportsbooks. Proposed solutions include a fixed fee only on profits, using a native POLY pool ...

Original Author:shtanga0x&securezer0

Compiled by | Odaily Planet Daily (@OdailyChina)

Translator | Wenser (@wenser2010)

Editor's Note: Recently, posts about Polymarket's LP incentives for the NCAA "March Madness" have almost flooded the timeline on platform X. At the same time, official Polymarket members hinted at a major announcement next Monday, with community speculation pointing towards funding or a token launch.

After the U.S. SEC and CFTC cleared the obstacle for airdrops on crypto platforms using the Howey test, POLY has become the "last hope for airdrop farming" in the eyes of many, and LP market making might become one of the key metrics for the airdrop.

In light of this, Odaily Planet Daily will use the opposing views of two analysts on LP market making incentives to provide Polymarket users with a more comprehensive perspective. The following is the compiled content, with some information edited.

Positive View: The 4 Major Categories Behind Polymarket's LP Incentive Program

Recently, Polymarket's incentive mechanism underwent a low-key upgrade, shifting its focus to Liquidity Providers (LPs). For the past few years, the platform has maintained a "zero trading fee" strategy, but since the beginning of this year, it has quietly introduced fees for specific prediction markets and launched two major market maker reward programs.

On the surface, charging trading fees seems unfavorable for trading users, but in reality, it addresses the core structural pain point of prediction markets—the liquidity problem.

The new fee structure aims to fund incentive programs, rewarding users who provide limit orders and maintain order book depth. Therefore, both Polymarket and its users benefit from: narrower spreads, richer order books, and a better trading experience—especially in high-frequency crypto markets.

Its rollout path is also very clear, showing a trend from single to multiple:

  • January 2026: 15-minute Crypto markets
  • February 2026: Expanded to 5-minute Crypto markets + NCAAB college basketball + Serie A football
  • March 6, 2026: Expanded to all Crypto markets (covering 1H, 4H, daily, weekly events, etc.)

Based on the above information, this article will detail how the new fee and reward system works—and why fees paid + rewards earned could become a potential anti-sybil indicator in the POLY airdrop. This is not a simple monetization operation; it's Polymarket showing through action that what it truly wants is liquidity, not volume-bot farming.

Part I. Complete Analysis of the New Taker Fee Mechanism

The vast majority of Polymarket markets remain completely free. Deposits, withdrawals, and trading (for most event markets) still have zero platform fees.

Trading fees currently only apply to the taker side (taker orders) and cover three types of markets:

  • All Crypto up/down markets (15min, 5min, 1H, 4H, daily, weekly, etc.)
  • NCAAB (US College Basketball League)
  • Serie A (Italian Football League)

The key point is that taker fees only take effect for markets created after the fee activation date; existing prediction events created previously are not affected.

The fee formula is uniform (where C = number of shares traded, p = share price/market probability, fees rounded to 4 decimal places, minimum fee is 0.0001 USDC):

The effective fee rate follows a symmetric probability curve:

  • Fees are highest when the probability is near 50% (highest outcome uncertainty);
  • Fees approach 0 when the probability is near 0 or 1 (higher outcome certainty).

For example, for a $100 trade in a Crypto market:

  • p=0.50 → trading fee ≈ $0.44;
  • p=0.10 or 0.90 → trading fee ≈ $0.02.

The probability curve for sports events is similar, but the midpoint fee (around 50% probability) is slightly higher. The fee direction is specifically:

  • Buy: Fee deducted from the share quantity;
  • Sell: Fee deducted from the USDC funds;
  • Market making incentives are paid in USDC.

It is worth mentioning that the Polymarket platform does not keep the entire fee pool; a fixed percentage of the fees (20% for Crypto markets, 25% for sports prediction events) is directly returned to LPs. (Note: Polymarket's US compliant platform uses a simple fixed fee of 0.01%. This analysis only discusses the global CLOB platform, which introduced the new fee system in 2026.)

Part II. Market Maker Incentive Program (Limit Order Execution Rewards)

Incentives in this section only cover markets where taker fees are collected. This means only limit orders that are executed (taken) by traders are eligible for corresponding rewards; simply placing an order without execution does not count.

The reward amount uses the same calculation method as the taker fee. Each participant's reward is proportional to their trading volume, and the total prize pool consists of a portion of the collected fees (20% for Crypto markets, 25% for sports prediction events).

Competition occurs only within specific prediction events; LP orders only compete with other LPs in the same liquidity pool for that event.

Daily incentives are sent directly in USDC to the corresponding wallet address.

Part III. Liquidity Incentives (Resting Order Incentives)

The second incentive system is provided by the Polymarket platform and applies to all prediction events (including those without fees).

The core difference is: Order execution is NOT required; you can earn money just by placing orders on the order book to provide liquidity.

Each prediction event defines several parameters that determine eligibility:

  • Maximum incentive spread (e.g., ±4 cents)
  • Minimum order size
  • Daily total reward pool

The platform samples the order book every minute, recording 10,080 snapshots per week.

The reward calculation formula is highly detailed:

1. Distance Score (Quadratic Equation)

Where,

V - Maximum incentive spread

s - Distance from the midpoint

Orders closer to the midpoint score exponentially higher.

2. Two-Sided Score (YES/NO Complementary Structure)

Bid and ask orders are scored separately, considering the complementary structure of Yes/No markets.

3. Q-value Minimization Adjustment

Providing liquidity on both sides of the order book scores higher points for the prediction event.

One-sided quoting is penalized unless the market probability is near 0 or 1.

4. Final Score

All LP scores are normalized and aggregated over time to determine each participant's proportional share of the market's reward pool.

Rewards are distributed in USDC at UTC midnight, with a minimum payout of $1.

Active reward prediction events and personal earnings can be viewed in real-time at polymarket.com/rewards. The incentive spread range is highlighted in blue on the order book interface. Users can also check the Polymarket official documentation.

Currently, one-sided orders can still earn points (but at a significant discount), while two-sided quoting is prioritized for incentive points. Rewards are calculated separately for each individual prediction event. There is no cross-event calculation. Effectively, the system rewards traders who maintain tight spreads and balanced liquidity near the market midpoint, enhancing the trading experience for all users.

Part IV. Sponsored LP Incentives

The third mechanism allows anyone to directly add LP incentives to specific markets using USDC, attracting LPs to provide market making. Sponsors can add or withdraw funds at any time; unused funds are automatically returned.

The rules of this mechanism are completely consistent with the Liquidity Incentive Program—just place orders, no execution required.

A typical case is the prediction event "Will Jesus Christ return by 2027?". A platform user invested $70,000 as an LP incentive in February and now receives about $57 per day in liquidity incentives, making this event one of the deepest markets on the platform. This mechanism allows the community itself to deeply activate market liquidity for any prediction event without waiting for official Polymarket action.

Part V. The Strongest Anti-Sybil Indicator for the POLY Airdrop

At first glance, Polymarket seems to just need more traders.

However, if most users rely solely on market orders, the platform will quickly face liquidity problems.

Polymarket does not rely on centralized market makers, so if limit orders are insufficient, the order book becomes sparse.

In this situation, it is difficult to avoid high slippage and sudden fee increases when buying, selling, or executing large orders.

Polymarket doesn't need volume-bots; it needs LPs who provide real value.

Previously, everyone only focused on trading volume, thinking high volume was the key to getting the airdrop. However, the new fee structure and reward program suggest a different incentive model—what matters is not just volume, but participation in fee-generating prediction events that need liquidity. In other words, the platform rewards targeted LP activity, not just passive limit orders.

The reward distribution formula effectively reveals the type of liquidity Polymarket values most. The scoring system evaluates:

  • Proximity of the order to the midpoint
  • Order size
  • Balance between bid and ask prices

Therefore, rewards become a direct measure of a trader's value to the platform's liquidity. If a trader consistently earns rewards, it means their orders are actively improving market liquidity and execution quality. Here are examples of potential incentives for market participants:

  • Will Arctic sea ice reach its maximum extent this winter?—Existed for 3 months, volume less than $20k, liquidity reward only $9;
  • Will Bitcoin reach $75,000 in March?—Existed for two weeks, volume reached $3 million, liquidity reward $142;
  • Bitcoin Up/Down - 15 minutes Covers hundreds of prediction events, daily trading volume millions of dollars, average daily fees ~$10k, liquidity reward $2000.

More important than specific prediction events is the truth revealed by the data—taker fees and earned liquidity rewards are harder to manipulate artificially compared to simple volume metrics. Systematically earning market making incentives requires capital, risk management, and constant presence, which significantly削弱 (weakens) the advantage of airdrop farmers and instead benefits genuine market participants.

Conclusion: Taker Fees and LP Incentives May Become Key Metrics for POLY Airdrop

Future POLY token distribution may depend not only on trading volume but also on taker fees paid and LP rewards earned. These metrics are transparent, measurable, and highly aligned with platform needs. In this model, rewards are not tied to farming volume but to contributions that truly optimize the platform's trading experience: liquidity, stability, and efficient price discovery.

In other words, the best performing LPs are the most valuable users. The most hardcore Polymarket players were never the ones with the highest trading volume, but the LPs who provide the deepest order book liquidity.

Also attached is a Polymarket LP market making guide: 《Now is the Best Time to Interact with Polymarket (Exclusive Tutorial)》

Of course, there are always different views in the market. Some believe that Polymarket's LP incentive program appears to be "spending money to buy liquidity" but is actually profiting from it, setting a trap for LP users. Let's hear the opposing view.

Opposing View: Is Polymarket's LP Incentive a Platform Scam? Are LPs Actually a "Pay-to-Lose" Trap?

Regarding Polymarket's recently launched LP incentive program, arbitrage trader and Polymarket/Kalshi bot player securezer0 directly called the much-hyped "Polymarket Rewards Farming" within the circle a massive psychological warfare operation, pointing out that it is a collective hype campaign paid for directly by the platform or heavily incentivized KOLs.

The Truth for LPs: Another Form of "Paying to Lose Money"?

Multiple LPs stated bluntly: The current LP mechanism on Polymarket is essentially "spending money to incur losses".

Where is the problem? The leaderboard directly incorporates LP rewards into the P&L data display but remains silent on a key concept—LP wear and tear (impermanent loss).

When your position gets filled on one side, it often cannot be sold at a reasonable price, or cannot be sold at all before the event resolves. This part of the capital loss is systematically hidden by the platform. The real ROI data is much lower than the book value. For most LP participants, the profit is actually negative. They just believe the POLY airdrop will cover the losses—this is not an arbitrage incentive program; it's a trade based on platform faith.

Why Are Professional Market Makers Reluctant to Enter?

Professional market makers generally avoid Polymarket LP market making for one core reason: Insider trading risk is real.

Both Polymarket and Kalshi had to offer equity in exchange for liquidity to get professional market makers to the table—this in itself speaks volumes.

Effective LP operation is an extremely complex automated risk control system. The myth of "low barrier to entry, high returns" for LPs only holds true if Polymarket continues to heavily subsidize liquidity rewards—a path that is fundamentally unsustainable in the long run.

The Platform's Real Dilemma: Needing to "Create" Millions of Dollars Out of Thin Air Daily

Insufficient liquidity is the primary driver behind Polymarket gradually turning on the fee switch.

To maintain liquidity rewards for various prediction events and keep more USDC liquidity on the platform, Polymarket needs to consume millions of dollars daily to maintain trading depth. If a better solution cannot be found, the platform has no choice but to charge fees on every buy and sell, using this revenue to support investors and market makers.

Once fees are fully implemented, ordinary users will be in an extremely awkward position—because traditional sports betting platforms might actually be more cost-effective, for the following reasons:

  • Odds are comparable after factoring in fees;
  • Traditional platforms offer cashback and monetary incentives;
  • Rules are clear and regulated;
  • Insider traders face account bans and even prison risks.

3 Viable Solutions: Fixed Fees on Profit, POLY Liquidity Pools, Extended Product Fees

Instead of seeking a quick fix that worsens the situation, it's better to address the root cause: Target the vampires, not the users. Charge fees on the arbitrage bots that extract USDC from real users. After all, these bots are the source of polluted liquidity. Specifically, the methods are:

Charge a fixed 1% fee only on profits, i.e., only on the net gain (sell price minus cost basis), not touching the principal, thus not harming the user trading experience.

Build a native liquidity pool with POLY tokens. Provide liquidity programmatically at the protocol level for various prediction events, deeply binding token economics with liquidity provision.

Charge on extended products, not the core product. Parlays, derivatives, leverage—these are natural fee scenarios; touching these won't damage the fundamental user experience.

Currently, the moat of Polymarket's platform business still needs strengthening. Zero fees and better odds are its most important value anchors区别于 (differentiating it from) traditional betting platforms. Abandoning these two points for short-term revenue is tantamount to self-sabotage.

Pertanyaan Terkait

QWhat are the four main categories of Polymarket's LP incentive programs as described in the article?

AThe four main categories are: 1. The new taker fee mechanism, 2. The market maker incentive program (limit order execution rewards), 3. The liquidity incentive (idle order rewards), and 4. The sponsored LP incentive.

QAccording to the opposing view, why might professional market makers be hesitant to participate in Polymarket's LP program?

AProfessional market makers are hesitant due to the real risk of insider trading. The article states that Polymarket and Kalshi had to offer equity in exchange for liquidity to attract professional market makers, indicating the inherent challenges and risks in the market structure.

QWhat is the core structural problem that Polymarket's new fee structure aims to solve?

AThe new fee structure aims to solve the core structural problem of liquidity in prediction markets. The fees are designed to fund incentive programs that reward users who provide limit orders and maintain order book depth.

QWhat does the opposing view claim is the true nature of being an LP on Polymarket, despite the visible rewards?

AThe opposing view claims that being an LP is essentially 'paying to lose money.' The visible rewards on leaderboards mask the critical concept of 'LP wear and tear' or losses incurred when positions are filled on one side and cannot be sold at a reasonable price before market resolution, often leading to a negative real ROI.

QWhat alternative solutions does the opposing view propose for Polymarket instead of charging taker fees?

AThe opposing view proposes three alternative solutions: 1. Charge a fixed 1% fee only on profits (net gains from selling price minus cost basis), 2. Build a native liquidity pool with the POLY token to programmatically provide liquidity, and 3. Charge fees on extended products like parlays, derivatives, and leverage instead of the core betting markets.

Bacaan Terkait

Setelah GPU dan Penyimpanan: MLCC Sedang Menjadi Perintis Pasar Triliunan Dolar Selanjutnya dalam Kekuatan Komputasi AI

**Ringkasan: MLCC – Pilar Tersembunyi di Balik Revolusi AI** Ketika fokus dunia tertuju pada GPU dan memori dalam perlombaan kekuatan komputasi AI, sebuah komponen kecil namun kritis justru mulai menanjak harganya: **Multilayer Ceramic Capacitor (MLCC)**. Dari komoditas murah, MLCC kini berubah menjadi bahan strategis. **Ledakan Permintaan dari AI dan Kendaraan Listrik:** * **Server AI** adalah penggerak utama. Dibanding server tradisional yang memakai ~2.000 MLCC, server pelatihan 8-GPU butuh 25.000-28.000 unit. Platform generasi baru seperti GB300 NVL72 butuh **440.000 unit**, dan Vera Rubin bisa mencapai **600.000 unit**. * Kebutuhan meledak karena GPU makin bertenaga dengan voltase rendah (<1V) dan arus tinggi (hingga 1.800A). MLCC berfungsi sebagai "penstabil" untuk menjamin daya listrik yang bersih dan stabil untuk chip. * **Kendaraan listrik** dan kendaraan otonom tingkat tinggi menjadi pilar kedua, dengan kebutuhan hingga 15.000-20.000 unit per mobil. **Pasokan Terbatas dengan Hambatan Tinggi:** Sisi pasokan sulit mengimbangi permintaan yang melesat 34% per tahun. Ekspansi kapasitas lambat (~10% per tahun) karena: 1. **Hambatan Teknologi & Modal:** Proses manufaktur sangat kompleks, membutuhkan mesin khusus dan material bubuk keramik mutakhir. Satu lini produksi canggih butuh investasi $3-5 miliar dan waktu 4-5 tahun untuk beroperasi penuh. 2. **Siklus Kualifikasi yang Panjang:** Kualifikasi untuk server AI butuh 12-18 bulan, sedangkan untuk otomotif 2-3 tahun, menciptakan keterikatan yang kuat dengan pemasok yang sudah ada. 3. **Disiplin Produsen:** Belajar dari siklus "boom-and-bust" sebelumnya, produsen utama kini lebih berhati-hati dalam menambah kapasitas untuk menjaga stabilitas harga jangka panjang. **Tiga Raksasa yang Mendominasi Pasar:** Pasar MLCC kelas tinggi dikuasai oleh tiga pemain utama: 1. **Murata (Jepang):** Pemimpin pasar dengan pangsa ~40% secara global dan 45-70% di segmen server AI. Pilihan utama untuk stabilitas dan kepastian. 2. **Samsung Electro-Mechanics (Korea Selatan):** Pemain dengan pertumbuhan dan ekspansi paling agresif, menawarkan potensi apresiasi tertinggi. 3. **Taiyo Yuden (Jepang):** Perusahaan dengan kemurnian eksposur tertinggi terhadap bisnis MLCC (~71% pendapatan), menjadikannya pilihan bagi yang menginginkan eksposur paling langsung ke sektor ini. **Kesimpulan:** MLCC, komponen yang selama ini diabaikan, telah menjadi garda depan dalam revolusi AI. Gabungan dari **permintaan eksponensial** (terutama dari server AI dan EV) dan **pasokan yang kaku** (karena hambatan teknologi dan disiplin industri) menciptakan kondisi untuk siklus super yang diperkirakan bertahan hingga 2030. Saat komputasi menjadi "minyak" era baru, MLCC adalah "pipa" vital yang mengatur setiap tetes listriknya.

marsbit26m yang lalu

Setelah GPU dan Penyimpanan: MLCC Sedang Menjadi Perintis Pasar Triliunan Dolar Selanjutnya dalam Kekuatan Komputasi AI

marsbit26m yang lalu

Baru saja, Claude Mythos 5 Rilis, 50 Juta Baris Kode Selesai dalam 1 Hari

Anthropic secara resmi meluncurkan model AI terkuat mereka, Claude Mythos 5, bersama versi aman bernama Claude Fable 5. Fable 5 tersedia untuk umum namun dilengkapi sistem keamanan yang secara otomatis menurunkan ke model Opus 4.8 jika mendeteksi permintaan berisiko tinggi seperti pembuatan malware atau riset biologi/kimia tertentu. Sementara itu, Mythos 5 yang "versi lengkap" hanya tersedia bagi pengguna terpercaya dengan kemampuan tak terbatas di bidang seperti keamanan siber dan penelitian ilmiah. Secara teknis, Fable 5 menunjukkan kemampuan luar biasa. Dalam pengujian SWE-bench Pro, ia mencetak 80.3%, mengungguli kompetitor. Kasus nyata dari Stripe menunjukkan Fable 5 dapat bermigrasi seluruh basis kode Ruby sebesar 50 juta baris hanya dalam 1 hari, pekerjaan yang biasanya membutuhkan tim engineering dua bulan. Model ini juga unggul dalam pemahaman visual, mampu menyelesaikan permainan Pokemon tanpa alat bantu, dan dalam tugas analisis finansial serta hukum. Di bidang penelitian, Mythos 5 menunjukkan lompatan besar. Model ini dapat menjalankan seluruh alur kerja ahli biologi secara mandiri, merancang senyawa protein, dengan 9 dari 14 desainnya sudah masuk pipeline pengembangan obat nyata. Dalam satu kasus, model kecil yang dilatih mandiri oleh Mythos 5 bahkan mengungguli penelitian terbaru yang diterbitkan di jurnal *Science*. Anthropic juga memperkenalkan perubahan paradigma keamanan, di mana kemampuan dan keamanan dipisahkan melalui sistem "klasifikasi dan penurunan model". Harga API untuk kedua model ini ditetapkan $10 per juta token input dan $50 per juta token output. Pengalaman pengguna awal, seperti dari profesor Ethan Mollick, mengindikasikan pergeseran pola kolaborasi manusia-AI. Pengguna merasa lebih seperti "klien" atau "pemberi kerja" yang memberikan tugas kompleks, sementara AI beroperasi secara mandiri seperti "studio" kecil yang menangani perencanaan dan eksekusi detail, menghasilkan produk akhir yang siap pakai.

marsbit1j yang lalu

Baru saja, Claude Mythos 5 Rilis, 50 Juta Baris Kode Selesai dalam 1 Hari

marsbit1j yang lalu

Yang Pertama Membawa AI OS ke 1,4 Miliar Orang, Ternyata WeChat?

AI WeChat Akhirnya Bergerak. Pada hari yang sama dengan WWDC Apple, WeChat merilis panduan bagi pengembang untuk mengintegrasikan aplikasi mini ke dalam ekosistem AI-nya. Pengembang kini dapat mengizinkan AI WeChat untuk membaca, mengoperasikan, dan memanggil fungsi aplikasi mini mereka. WeChat menawarkan dua mode integrasi: "Mode Otomatis" yang memungkinkan AI menganalisis kode dan halaman aplikasi mini tanpa coding tambahan, dan "Mode Pengembangan" untuk membuat Skill yang dikustomisasi. Langkah ini mengubah seluruh ekosistem WeChat—jutaan aplikasi mini, WeChat Pay, notifikasi layanan—menjadi lapisan eksekusi untuk AI. Dibandingkan dengan pendekatan Apple Siri yang perlu bernegosiasi dengan setiap aplikasi pihak ketiga, arsitektur terpusat WeChat memberinya "sudut pandang Tuhan." Setiap kode aplikasi mini melewati sistem WeChat, memungkinkan analisis otomatis dan penerjemahan menjadi alat yang dapat dipanggil AI. Keunggulan infrastruktur ini tidak dimiliki oleh Apple atau Google. Dengan 1,432 miliar pengguna aktif bulanan dan cakupan layanan harian yang luas melalui aplikasi mini, WeChat berpotensi menjadi sistem operasi yang dioperasikan dengan bahasa alami. Pengguna dapat memberi perintah seperti "pesan tiket kereta untuk besok," dan AI akan menjalankan tugas melalui aplikasi mini yang relevan. Meski tantangan seperti akurasi dan kepercayaan dalam transaksi tetap ada, WeChat telah membangun jaringan layanan yang matang. Pencapaian sejati AI Agent adalah penyelesaian tugas secara "tanpa disadari" oleh pengguna, dan WeChat berada paling dekat untuk mewujudkannya.

marsbit1j yang lalu

Yang Pertama Membawa AI OS ke 1,4 Miliar Orang, Ternyata WeChat?

marsbit1j yang lalu

Apple dengan Kapitalisasi Pasar 4 Triliun Dolar, Mengapa Tak Bisa Menghadirkan Siri yang Cerdas?

Penulis: Hu Shixin, Editor: Ye Jinyan Pada acara WWDC2026, Apple meluncurkan Apple Intelligence dan Siri AI yang diperbarui, menjawab pertanyaan tentang kemampuan AI perusahaan yang tertinggal. Siri AI yang baru didukung oleh teknologi model dasar dari kolaborasi dengan Google (Gemini), tetapi dijalankan pada model Apple sendiri yang dioptimalkan untuk perangkatnya. Siri kini dapat memahami konteks layar, data pribadi, dan menjalankan tugas di berbagai aplikasi, berubah dari asisten suara menjadi pusat operasi sistem. Namun, kemampuan yang ditunjukkan sebagian besar sudah umum di industri, dan fungsinya belum tersedia di UE dan Tiongkok daratan karena regulasi. Peluncuran ini dianggap sebagai upaya Apple untuk mengejar ketertinggalan setelah lambat merespons revolusi AI generatif, yang dipicu oleh kesuksesan ChatGPT. Artikel ini menelusuri "utang AI" Apple di era CEO Tim Cook, termasuk perkembangan Siri yang lambat sejak akuisisi tahun 2010, pergantian kepemimpinan yang sering, dan keterlambatan dalam pengembangan model dasar. Proyek Apple Intelligence sempat tertunda pada 2024 karena masalah stabilitas, mendorong reorganisasi tim dan kolaborasi dengan Google. Analis berpendapat bahwa tujuan utama Apple bukanlah menciptakan model terkuat, melainkan mempertahankan Siri sebagai pintu masuk utama di perangkatnya untuk mengontrol konteks pribadi pengguna dan potensi pendapatan masa depan. Masa depan pertumbuhan Apple bergantung pada kemampuannya membuktikan bahwa AI dapat mendorong siklus pergantian perangkat atau pendapatan layanan baru, sambil mengatasi tantangan pengembangan perangkat keras alternatif seperti kacamata pintar.

marsbit1j yang lalu

Apple dengan Kapitalisasi Pasar 4 Triliun Dolar, Mengapa Tak Bisa Menghadirkan Siri yang Cerdas?

marsbit1j yang lalu

Trading

Spot
Futures

Artikel Populer

Cara Membeli 4

Selamat datang di HTX.com! Kami telah membuat pembelian 4 (4) menjadi mudah dan nyaman. Ikuti panduan langkah demi langkah kami untuk memulai perjalanan kripto Anda.Langkah 1: Buat Akun HTX AndaGunakan alamat email atau nomor ponsel Anda untuk mendaftar akun gratis di HTX. Rasakan perjalanan pendaftaran yang mudah dan buka semua fitur.Dapatkan Akun SayaLangkah 2: Buka Beli Kripto, lalu Pilih Metode Pembayaran AndaKartu Kredit/Debit: Gunakan Visa atau Mastercard Anda untuk membeli 4 (4) secara instan.Saldo: Gunakan dana dari saldo akun HTX Anda untuk melakukan trading dengan lancar.Pihak Ketiga: Kami telah menambahkan metode pembayaran populer seperti Google Pay dan Apple Pay untuk meningkatkan kenyamanan.P2P: Lakukan trading langsung dengan pengguna lain di HTX.Over-the-Counter (OTC): Kami menawarkan layanan yang dibuat khusus dan kurs yang kompetitif bagi para trader.Langkah 3: Simpan 4 (4) AndaSetelah melakukan pembelian, simpan 4 (4) di akun HTX Anda. Selain itu, Anda dapat mengirimkannya ke tempat lain melalui transfer blockchain atau menggunakannya untuk memperdagangkan mata uang kripto lainnya.Langkah 4: Lakukan trading 4 (4)Lakukan trading 4 (4) dengan mudah di pasar spot HTX. Cukup akses akun Anda, pilih pasangan perdagangan, jalankan trading, lalu pantau secara real-time. Kami menawarkan pengalaman yang ramah pengguna baik untuk pemula maupun trader berpengalaman.

748 Total TayanganDipublikasikan pada 2025.10.20Diperbarui pada 2026.06.02

Cara Membeli 4

Diskusi

Selamat datang di Komunitas HTX. Di sini, Anda bisa terus mendapatkan informasi terbaru tentang perkembangan platform terkini dan mendapatkan akses ke wawasan pasar profesional. Pendapat pengguna mengenai harga 4 (4) disajikan di bawah ini.

活动图片