Solana’s Firedancer Now Charges 1 SOL Per Bug Report
Solana’s Firedancer workforce now requires researchers to burn 1 SOL per bug report back to fight AI-generated submissions flooding its safety program.
TLDR
- Firedancer, the Solana validator consumer constructed by Jump Crypto, now requires researchers to burn 1 SOL (about $100) per bug report.
- The price is nonrefundable, even for invalid, out-of-scope or duplicate reviews.
- A latest audit contest drew 809 reviews however produced solely 127 distinctive safety findings, and Jump believes AI helped create almost all of them.
- Jump is briefly boosting rewards, with important Alpenglow flaws eligible for as much as $300,000.
The team behind Solana’s Firedancer software has launched a paywall for bug hunters. Researchers should now burn 1 SOL, roughly $100, with each submission. The aim is to chop down on AI-generated spam and provides engineers extra time to analyze actual safety flaws.
Under phrases up to date on Monday, the burned tokens are completely faraway from circulation. The price is nonrefundable whether or not a report seems to be invalid, outdoors this system’s scope, or a replica. Reports submitted and not using a burn will not be eligible for rewards.
What Is Firedancer and Why Does It Matter?
Jump Crypto constructed Firedancer as an alternative choice to Agave, the principle software program validators use to verify and course of Solana transactions. A severe bug on this software program can cease validators from processing funds or trigger them to disagree about account balances. That might disrupt trades and transfers throughout the community. Every hour spent sorting by means of duplicate or unworkable reviews is an hour not spent fixing vulnerabilities like these.
AI Floods the Bug Bounty Program
The resolution follows a $1 million audit competitors that ran from April 9 to May 9. It attracted 809 reviews from 162 researchers, but produced solely 127 distinctive safety findings as soon as non-exploitable observations have been excluded. Nearly half of the legitimate submissions have been duplicates.
Jump believes AI brokers helped discover or write nearly each submission. In an Oct. 16 assessment, Jump’s Ishan Bhatt argued that AI has made discovering bugs low-cost whereas judging them stays costly. Engineers nonetheless needed to reproduce every drawback and verify whether or not the proposed assault might really work. Some reviews flagged obvious weaknesses whereas overlooking protections elsewhere within the code.
The assessment backlog peaked at 299 open reviews. As checks slowed, different researchers rediscovered bugs that had been submitted however not but mounted or disclosed. About 80% of duplicates arrived greater than three days after the unique report.
The contest had already charged $100 per submission and capped researchers at one report a day. Jump says charging a price discourages individuals from sending giant volumes of speculative findings. The new platform additionally limits every predominant report part to three,000 characters.
Higher Rewards for Alpenglow Bugs
Alongside the brand new price, Jump is elevating payouts. Eligible findings obtain double the usual reward by means of Nov. 20, and triple for bugs particular to (*1*). Critical Alpenglow findings can qualify for as much as $300,000.
The transfer exhibits how AI is altering safety work in crypto. Automated instruments can generate reviews sooner than people can assessment them, so groups are actually utilizing financial prices to filter the noise.
