Solana price outlook tied to Fed policy and US-Iran tensions
Solana faces selling pressure due to hawkish Fed and US-Iran tensions; recovery may require dovish pivot or geopolitical breakthrough.
Binance hack highlights crypto risks; AI aids attacks but scarce digital assets remain valuable.
On October 29, attackers stole approximately $6.3 billion from Binance, marking the largest heist in crypto history and nearly four times the prior record. The breach originated months earlier when an engineer searched for an AI coding tool.
In the aftermath, something occurred that I hadn't witnessed since Bitcoin's all-time high in October 2025. People outside the crypto space started asking me what was happening.
I believe insiders haven't fully grasped this. We typically measure events by their impact on prices and withdrawals.
A more relevant metric is whether the event escapes our corner of the internet, and this one did. For many, it is the first news they've heard about crypto in quite some time.
This makes it a fitting moment for a fundamental question. AI is transforming how software is written, markets are traded, and attacks are executed. What role does crypto play in that context? The answer starts with bad news and improves from there.
Binance has detailed how the attack unfolded. In June, a wallet engineer searching for an AI tool installed a compromised package instead.
In September, another engineer participated in fake job interviews via live video and received a coding assignment that contained malware unseen by any security product.
From that engineer's machine, the attackers used one of Binance's internal AI tools to compromise two more engineers and then the wallet-signing service. The FBI has attributed the theft to North Korea.
No evidence shows that AI planned or executed this attack, and I won't assert that. However, AI served as the initial bait and a tool in the middle, and attackers are adopting it rapidly. TRM Labs reported a record 201 crypto hacks in the first half of 2026, compared to 83 a year earlier.
We have also been on the receiving end. Phemex suffered a breach in January 2025, prompting a complete rebuild of our security architecture.
Binance states it has replaced nearly all affected assets from its own funds, with an independent review pending.
Investigators at Elliptic later discovered the thieves creating their own meme tokens to launder part of the funds. It was a small portion of the total, but it reflects how this industry appears from the outside: meme coins on one side and exploits on the other.
If that's all crypto amounts to in a few years, little of the industry will remain. I don't believe that's where we're headed, and AI is a big reason why.
AI makes digital things abundant. Code, apps, images, music, and soon entire films can be created by anyone in minutes. When something becomes that easy to produce, its value trends toward zero.
Bitcoin is the opposite type of digital object. Ask a model for a thousand logos, and you get them in a minute. Ask it for one bitcoin, and it must buy it like anyone else. No model can generate another coin or alter the ledger that records ownership.
This extends beyond Bitcoin. As generated images and video fill the internet, people will seek digital goods that are scarce enough to retain value, along with records of photos or videos that can't be tampered with afterward.
NFTs and other blockchains can offer both. Much of the NFT crowd has moved on to meme coins or physical collectibles, which address scarcity differently, but the underlying question remains.
AI agents are also part of this picture. As I argued in February, they need money they can hold and move without bank approval, and crypto has already built that. Scarce assets, immutable records, and open rails are what this industry offers that AI cannot replicate.
Recent security incidents across crypto are a reminder that protecting user assets requires constant vigilance.
— Phemex (@Phemex_official) November 12, 2026
At Phemex, security goes beyond wallets. Our approach combines multiple layers of protection across people, devices, systems, and transactions.
Here's how. 🧵 pic.twitter.com/OqeZF8CD3J
That's the long-term perspective. The practical one, for anyone with an exchange account today, is what AI does for their trading.
A year ago, a research lab called Nof1 gave six leading AI models $10,000 each to trade live crypto perpetuals with no human intervention. Four of the six lost money, with the worst losing about 60% in just over two weeks.
Prediction markets provide another data point, this time with real users. A Bloomberg analysis of roughly two million Polymarket wallets found that retail traders picked the correct outcome more often than bots did but still lost, because they entered late at worse prices.
I see that as positive news. The judgment was there, but timing and discipline were missing, areas where software excels.
That doesn't mean a bot should handle the trading. Most people who envision an AI trading for them imagine a static market while it operates. The real market will be full of other agents, each running competing strategies that adapt as conditions change.
The same Polymarket data shows the result: most bots lost money too, with profits concentrated in about 800 accounts.
Phemex also launched an AI Bot in February this year. It automates grid strategies and adjusts leverage as volatility changes.
Then there's another AI Agent we are about to launch that works differently. It explains what's moving the market and shows its evidence, converts a trading idea into an order draft, and reviews positions, PnL, and funding fees with the user afterward.
Last week we spoke about Phemex becoming AI-native at the architectural level, this week, we are moving to execution. Programmable and intuitive trading bots are the first operational step: embedding intelligent systems directly into live trading flows rather than keeping AI as a… pic.twitter.com/CxuMNGgoNd
— Federico0x @Phemex (@Federico0x) February 25, 2026
In such a market, an AI-assisted strategy wins in one of two ways. Either the technology behind it is superior to everyone else's, which is unlikely for a retail user, or the person directing it makes better decisions.
The second path is available to anyone, and that's where AI proves most valuable.
If you let any AI tool near your account, check three things first: that you can see why it took each trade, that it works within limits you set, and that you can stop it immediately.
Share to
Disclaimer: this article comes from third-party media and is provided for reference only. It does not constitute investment advice. Crypto and other financial products carry significant price volatility risk, so please make your own decisions carefully.
Solana faces selling pressure due to hawkish Fed and US-Iran tensions; recovery may require dovish pivot or geopolitical breakthrough.
US Dogecoin ETFs have attracted no new money since September 29, while some wallets continued buying during the price drop.
Bitwise projects Solana could reach $1,100 by 2031 if tokenized equities reach 5% of global stocks and Solana captures 65% of that trading.
Security researcher Scott Helme identified OPTIMJACK malware on 186 high-value websites, stealing crypto and card data via compromised Optimizely tool.