Fetch.ai and NuNet, two protocols operating in the decentralized artificial intelligence sector, were exploited by the same attacker, according to reports from BeInCrypto and The Cryptonomist. The combined losses across both platforms reached approximately $2 million.
The exploit had an immediate and visible effect on NuNet’s native token, NTX. The token fell to an all-time low in the aftermath of the attack. The Cryptonomist reported the decline reached as much as 70%, though the exact scale of the price move varies slightly between reports.
Details on the precise attack vector used against Fetch.ai and NuNet have not been fully disclosed in available reporting. What is clear is that the same actor is believed to be responsible for both incidents, suggesting a coordinated or opportunistic strategy targeting related infrastructure within the decentralized AI space.
Fetch.ai and NuNet both operate in a niche corner of the crypto market that blends blockchain infrastructure with machine learning and autonomous agent technology. This sector has attracted significant investor interest over the past two years, as artificial intelligence narratives have driven capital into tokens perceived as exposed to AI growth. That popularity has also made these projects more visible targets.
Exploits affecting smaller-cap tokens like NTX tend to produce outsized price swings, since liquidity is often thinner than in larger, more established markets. A $2 million exploit can represent a small fraction of total value locked in a major DeFi protocol, but for a token with limited trading depth, the same dollar figure can trigger a much sharper price collapse. That dynamic appears consistent with the drop observed in NTX following the incident.
Neither Fetch.ai nor NuNet has issued a detailed public post-mortem covering the full scope of the exploit, based on currently available reporting. Investors and community members are likely watching for official statements from both teams addressing the cause of the breach, any funds recovery efforts, and plans to prevent similar incidents going forward.
Sources disagree on this story
This article was published before the reports below were compared. The reporting above stands; what follows is where the published accounts do not agree.
The Cryptonomist EN and AMBCrypto agree a single attacker hit Fetch.ai and NuNet for roughly $2 million combined, but give different figures for NTX's price crash, FET's decline, and the dollar values of the FET drain and NTX mint.
What all sources agree on
- The same attacker is linked to exploits on both Fetch.ai and NuNet, involving roughly $2 million in combined value.
- The attacker used a valid/obtained authorization signature to drain FET from Fetch.ai's converter contract in a single transaction.
- The same wallet cluster also received newly minted NTX tokens from NuNet.
- NTX suffered a much larger relative price decline than FET as a result of the exploit.
Where the reports disagree
1Extent of NTX's price crash and its all-time low
NTX crashed more than 70% in 24 hours to an all-time low of $0.000328
The price of NTX fell from approximately $0.00130000 to $0.00005559. This was a 95.7% drop in value over the course of 408.5 million unauthorized tokens entering circulation.
What would settle it: On-chain price/exchange data for NTX at the time of the exploit
2Size of FET's price decline
FET, by comparison, fell about 5% over the same window.
The attack pushed FET down to about $0.1711, a roughly 9% decline from its peak.
What would settle it: On-chain price/exchange data for FET at the time of the exploit
3Dollar value of FET drained from the converter contract
draining about $1.56 million in FET from Fetch.ai's TokenConversionManagerV3.
The attacker extracted 8.72 million FET worth about $1.54 million.
What would settle it: The on-chain transaction on TokenConversionManagerV3
4Dollar value of NTX minted to the attacker's wallet
Blockaid estimated that mint at roughly $452,000 worth of tokens.
Those tokens were worth roughly $463,000.
What would settle it: The on-chain minting transaction for NTX
What to make of it
Treat the broad narrative — one attacker, two protocols, roughly $2 million combined, NTX hit harder than FET — as established, but don't rely on either outlet's specific percentages or dollar figures until on-chain data settles the discrepancy.
Market Impact
The immediate market reaction centered on NTX, which recorded a sharp decline to an all-time low. Sudden exploits at protocols with tightly linked technology stacks can also raise concerns among holders of related tokens, even when those tokens were not directly affected. Fetch.ai's own token may see indirect scrutiny from traders reassessing risk across the decentralized AI category.
Broader implications for the sector depend on how quickly both teams respond with transparent disclosures. Security incidents involving shared attackers across multiple protocols often prompt closer inspection of shared codebases, dependencies, or integration points. Until further technical details emerge, the exploit is likely to weigh on sentiment toward smaller AI-linked tokens more than on the wider crypto market.
The Fetch.ai and NuNet exploits underscore the security risks facing smaller decentralized AI projects as they scale. Further clarity from both teams on the attack's root cause will likely shape investor confidence in the near term.
Frequently Asked Questions
What happened to Fetch.ai and NuNet?
Both protocols were reportedly exploited by the same attacker, with combined losses estimated at around $2 million, according to BeInCrypto and The Cryptonomist.
What happened to the NTX token price?
NuNet's native token, NTX, fell to an all-time low following the exploit. One report cited a decline of as much as 70%.
Is Fetch.ai's own token affected?
Reports focus primarily on NTX's price decline. Direct impact on Fetch.ai's token has not been specified in available reporting.
Has either project explained how the exploit occurred?
Detailed technical explanations from Fetch.ai or NuNet were not included in the available reporting at the time of publication.