Every song you've ever played through Apple Music, YouTube, or an iPhone was probably AAC. Not MP3. The file that actually hit your speakers was encoded in a format most people have never consciously chosen. Apple switched to AAC as their default in the early 2000s. YouTube followed. Spotify uses OGG Vorbis, which comes from the same generation of codecs. MP3 is everywhere in name, but the audio world quietly moved on years ago.
So what's actually different between the two formats, and does any of it matter for you?
The technical gap (and why it exists)
Both MP3 and AAC are lossy codecs. They compress audio by throwing away information your ear is unlikely to notice: sounds masked by louder sounds, frequencies at the edges of human hearing, subtle spatial cues. This is called perceptual coding, and it's the same basic idea in both formats.
The gap comes from when they were built. MP3 dates from the early 1990s. AAC was standardized in 1997 specifically to replace it, with a more sophisticated compression model. At the same bitrate, AAC discards less of what matters.
The practical result, backed by years of double-blind listening tests: 128 kbps AAC sounds roughly equivalent to 160-192 kbps MP3. That's not a marginal improvement. You're getting the same perceived quality out of a significantly smaller file, or noticeably better quality at the same file size.
Where the difference is most audible: high-frequency content. Cymbal hits, acoustic guitar, vocals with sharp "s" sounds. At low bitrates, MP3 tends to smear these into something slightly underwater. AAC handles them more cleanly. At 320 kbps (MP3's ceiling), the two formats converge and most people can't reliably tell them apart in blind tests.
Does the bitrate table actually help you decide?
Here's a rough orientation:
| Bitrate | MP3 | AAC |
|---|---|---|
| 128 kbps | Acceptable, artifacts audible on good headphones | Good, clean enough for most listening |
| 192 kbps | Good | Very good |
| 256 kbps | Very good | Excellent (Apple Music's streaming tier) |
| 320 kbps | Excellent | Excellent |
Apple Music streams at 256 kbps AAC. YouTube uses AAC for its audio tracks. These choices were made by engineering teams with access to extensive listening tests. They picked AAC because at the same bitrate, it performed better.
The compatibility question
This is where MP3 still has a real argument. MP3 plays everywhere, full stop. Every device, every software player, every car stereo made in the last three decades. AAC has broad support on modern hardware but there are still edge cases: some older car stereos, basic portable players sold as "MP3 players" that take the name literally, industrial or embedded systems running old firmware.
The gap has narrowed considerably. Anything manufactured after roughly 2010 almost certainly supports AAC. But if you're distributing audio and have no idea what will play it, MP3 removes the guessing.
When the choice actually matters
For most people, the format question comes up in a few specific situations.
If you're building a personal music library, AAC from a lossless source gives you better quality at smaller file sizes than MP3. If your originals are already in FLAC or WAV, encoding to AAC at 256 kbps is a reasonable archival strategy for a listening copy.
If you're submitting audio to a podcast host, platform, or client, check what they actually accept. Some platforms specify MP3. In that case, you're not choosing.
If you're uploading to YouTube, the platform will re-encode your audio regardless of what you send. Uploading high-bitrate AAC or MP3 is fine; the output will be AAC at a bitrate YouTube controls.
If you're encoding audio for a web project, AAC has essentially universal browser support now and produces smaller files than MP3 at equivalent quality.
The transcoding trap
One question that comes up constantly: should you convert your existing MP3 library to AAC to get better quality?
No. Converting MP3 to AAC is transcoding, compressing an already-compressed signal a second time. The output won't be better than the original MP3. It will be worse, because you're layering a new round of compression artifacts on top of the first. The only source worth encoding to AAC is lossless audio: FLAC, WAV, AIFF. If you only have MP3s, leave them as MP3s.
Why MP3 is still the default assumption
AAC is technically better by almost any measure, yet MP3 remains the format people reach for by default. Part of this is inertia: MP3 became synonymous with "audio file" during the Napster era and that association stuck. Part of it is the name itself. People know what MP3 means. They don't always know what AAC is, even when they're listening to it.
The format you use matters less than the bitrate and the quality of the source. A 128 kbps MP3 ripped from a CD sounds better than a 256 kbps AAC encoded from a 128 kbps MP3. Start with the best source available, pick a reasonable bitrate, and the MP3 vs AAC question becomes secondary.
That said, if you're starting fresh with lossless audio and targeting modern devices, AAC is the better encoder. The streaming industry already made that call.